CYCLIC DISTILLATION FOR ENERGY CONSERVATION IN SPIRIT ...
Transcript of CYCLIC DISTILLATION FOR ENERGY CONSERVATION IN SPIRIT ...
CYCLIC DISTILLATION FOR ENERGY CONSERVATION IN SPIRIT PRODUCTION
By
Nicole Elizabeth Shriner
A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Chemical Engineering – Doctor of Philosophy
2018
ABSTRACT
CYCLIC DISTILLATION FOR ENERGY CONSERVATION IN SPIRIT PRODUCTION
By
Nicole Elizabeth Shriner
Distillation technology has been used worldwide throughout the chemical industry for
many years, originally being invented in China around 800 BC. There are currently over 40,000
columns in operation worldwide [29]. Although there are many variations in column types, sizes,
tray configurations, etc., the underlying principle is the same. Distillation is the physical process
of heating a liquid mixture to separate components based the relative volatilities of each
component. Distilleries consume about 40% of the total energy used to operate plants in the
chemical industry, and over 95% of that energy is used in separation processes [31]. Therefore,
improving distillation technology is one area that chemical engineers are working to transform
into a more efficient process by use of process intensification. The purpose of this research was
to investigate the possible economic and energy impacts of a form of process intensification
called ‘cyclic distillation’ as well as its application to the distilled spirits industry.
Cyclic distillation is an alternate mode of operation first proposed and studied by Cannon
and McWhirter in the 1960’s. The ‘cycle’ consists of alternating between two periods, vapor
flow and liquid flow. During the vapor flow period heat is supplied to the column, vapor flow
upwards and distillate is collected as in normal operation. During the liquid flow period, the heat
source is ceased and the liquid on each tray is transferred to the tray below. The cycle of
alternating between the two periods is continued for the entire distillation. Cyclic distillation has
been applied to many different systems and configurations and has been demonstrated
experimentally and theoretically to increase column throughput, lower energy requirements and
achieve higher separation performance.
In this study, a 150 L Carl © still was used to distill fermented apple cider and apple
brandy low wines. Cyclic distillation and conventional operation were compared using the same
system. Distillation samples were collected and analyzed using a gas chromatograph. The system
was also simulated using MATLAB.
In summary, cyclic distillation on the batch column used was able to show a decrease in
energy (steam) requirements for finishing runs but not in stripping runs. Cyclic distillation trends
included ethanol concentration decreased at a slower rate compared to conventional operation
and as a result temperature profiles mimicked this phenomenon. It was shown that the volume of
hearts (product) was increased and volume of tails and heads (unwanted by-products) was
decreased. This research has shown that the application of cyclic distillation on spirit production
is a viable option for distilleries large and small. In extension of this work, it is suggested that
cyclic distillation is applied to other types of spirits and to columns with a larger number of trays,
and to trays with true plug flow capability.
iv
This dissertation is dedicated to my mom, Dr. Tamara Shriner. Thank you for pushing me to
achieve my highest potential even when I was slightly opposed to continue my academic path.
Thank you for instilling in me that I can do anything I set my mind to. Thank you for showing
me how a hardworking and independent woman carries herself. I am forever indebted and
grateful to you.
I love you.
v
ACKNOWLEDGEMENTS
I would like to acknowledge and thank my advisor, Dr. Kris Berglund. Without him I would not
have found my passion in the fermented beverage industry or have the connections and
opportunities that I have and will continue to receive. His confidence in me since undergrad has
helped to instill confidence in myself and in my work. I would also like to thank my colleagues
Jacob Rochte, Yasheen Jadidi, and John Szfranski. Without their help I would likely still be
googling how to fix things in the lab, getting a lot less hours of sleep, and still working on errors
in my mat lab code. THANK YOU!
vi
TABLE OF CONTENTS
LIST OF TABLES ....................................................................................................................... viii
LIST OF FIGURES ....................................................................................................................... ix
KEY TO SYMBOLS .................................................................................................................. xvii
CHAPTER 1: Introduction ............................................................................................................. 1
1.1 Distillation Theory and Background ..................................................................................... 1 1.2 Distillation Improvement and Cyclic Distillation ................................................................. 2
1.3 Cyclic Distillation Application to Spirits Industry ................................................................ 3
CHAPTER 2: Literature Review .................................................................................................... 7
CHAPTER 3: Modeling Overview ............................................................................................... 30
3.1 Batch Distillation ................................................................................................................. 30
3.1.1 Alembic Distillation – The Rayleigh Equation ............................................................ 30 3.1.2 Multistage Batch Distillation ........................................................................................ 31 3.1.3 Stage-By-Stage Methods for Batch Rectification ........................................................ 34
3.1.4 Governing Equations and Thermodynamic Model for System in Study ..................... 38 3.2 Cyclic Continuous Distillation Modeling Approach ........................................................... 40
3.2.1 Assumptions ................................................................................................................. 40
3.2.2 Operational Constraints ................................................................................................ 40
3.2.3 Model of Vapor Flow Period ........................................................................................ 41 3.2.4 Model of Liquid Flow Period ....................................................................................... 41
3.2.5 Solution Method ........................................................................................................... 42 3.3 Design of Cyclic Distillation ............................................................................................... 43
3.3.1 Design Methodology .................................................................................................... 43
3.4 Batch Cyclic Distillation Dynamic Analysis ...................................................................... 45 3.4.2 Vapor Flow Period ....................................................................................................... 46 3.4.3 Liquid Flow Period ....................................................................................................... 48
3.4.4 Solution Method ........................................................................................................... 51
CHAPTER 4: Methods ................................................................................................................. 52
4.1 Materials and Equipment .................................................................................................... 52 4.1.1 Manufacturing Equipment ............................................................................................ 52
4.2 Measurement Methods ........................................................................................................ 53 4.2.1 GC Method ................................................................................................................... 53
4.3 Research Design .................................................................................................................. 55 4.4 Procedures ........................................................................................................................... 56 4.5 Data Collection and Analysis .............................................................................................. 56
CHAPTER 5: Results ................................................................................................................... 58
vii
5.1 Results ................................................................................................................................. 58
5.1.1 All Distillation Results ................................................................................................. 58
5.1.2 Component Concentration Results ............................................................................... 61 5.1.3 Results by Type of Distillation ..................................................................................... 65 5.1.4 Results of Cuts Taken ................................................................................................... 68 5.1.5 Temperature Profile Results ......................................................................................... 70
5.2 Simulation Results ............................................................................................................... 71
5.3 Reproducibility .................................................................................................................... 80
CHAPTER 6: Discussion .............................................................................................................. 81 6.1 Summary ............................................................................................................................. 81 6.2 Conclusions ......................................................................................................................... 81
6.2.1 All Distillation Discussion ........................................................................................... 81
6.2.2 Component Concentration Trends ................................................................................ 82 6.2.3 Low Wines vs Finishing Tends .................................................................................... 83
6.2.4 Effects on Cuts ............................................................................................................. 84
6.2.6 Reflux Observations and Discussion ............................................................................ 85 6.2.7 Simulation Discussion .................................................................................................. 85
6.3 Limitations and Sources of Error ........................................................................................ 86
6.4 Recommendations for Future Research and Application .................................................... 87
APPENDICES .............................................................................................................................. 88 APPENDIX A: Graphical Distillation Results .......................................................................... 89 APPENDIX B: MATLAB Codes ........................................................................................... 133
APPENDIX C: Miscellaneous Figures/Tables ....................................................................... 142
BIBLIOGRAPHY……………………………………………………………...……………………….143
viii
LIST OF TABLES
Table 2.1 Definition of Various Efficiencies [5] .......................................................................... 10
Table 2.2 Effects of Mixing During the Liquid Flow Period on the Separating Ability of a
Controlled Cycle Rectification Still at Total Reflux [5] ............................................................... 13
Table 2.3 Effect of Relative Volatility on the Overall Column Efficiency of a Controlled Cycling
Rectification Still at Total Reflux [5] ........................................................................................... 13
Table 3.1 Variables and Total Number of Equations for Multicomponent Batch Distillation ..... 37
Table 4.1 Detected Compounds [44] ............................................................................................ 54
Table 4.2 Distillation Conditions .................................................................................................. 55
Table 5.1 Distillation Efficiency Results ...................................................................................... 58
Table 5.2 Distillations Separated by Type .................................................................................... 65
Table 5.3 Cut Results for Distillations 11, 12, 13, 14 & 15.......................................................... 68
Table 5.4 Parameters for Batch Distillation Simulation ............................................................... 72
Table 5.5 Statistical Analysis of Reproducibility ......................................................................... 80
Table B.1 Cyclic Distillation Simulation Data ........................................................................... 141
ix
LIST OF FIGURES
Figure 2.1 Cycling Increases Column Capacity for Type-Two Plates [2] ...................................... 7
Figure 2.2 Compositions 𝒙𝒊 𝑽 as a Function of Fraction of Plate Holdup Dropped (𝝓) [6] .......... 9
Figure 2.3 Effective Plate Efficiency 𝑬𝟎 as a Function of Plate Number [6] .............................. 10
Figure 2.4 Effect of Fraction of Plate Holdup Dropped During the Liquid Flow Period on the
Over-all Efficiency of a Controlled Cycling Rectification Still at Total Reflux [5] .................... 12
Figure 2.5 Effective Plate Efficiencies (E0) in a Controlled Cycling Rectification Column Still at
Total Reflux [5]............................................................................................................................. 14
Figure 2.6 (a) Dependence of Column Separation Efficiency E on Mean Vapor Velocity 𝑾𝒗 (b)
Dependence of Column Separation Efficiency E on Vapor Feed Time 𝝉𝒗 and on Liquid-feed
Time 𝝉𝒍 (c) Dependence of Column Separation Efficiency E on the Ratio of Feed Periods of
Phases 𝝉𝒗/𝝉𝒍 for Various Velocities [13] ..................................................................................... 17
Figure 2.7 Schematic of Column Used for Stepwise Period Distillation [18] .............................. 19
Figure 2.8 New Tray Design for Period Cycling Distillation [23] ............................................... 21
Figure 2.9 The Three Characteristic Periods in the Cyclic Operation of a Regular Batch
Distillation Column [39] ............................................................................................................... 22
Figure 2.10 Lab Batch Column Used to Apply a Batch Cyclic Operating Theory [39] ............... 23
Figure 2.11 Batch Rectification Control Diagram [25] ................................................................ 26
Figure 2.12 Batch Stripping Control Diagram [25] ...................................................................... 26
Figure 2.13 Mass Exchange Contact Device and Column [28] .................................................... 29
Figure 3.1 Simple Batch Distillation Schematic ........................................................................... 30
Figure 3.2 Multistage Batch Distillation ....................................................................................... 32
Figure 3.3 McCabe-Thiele Diagram for Multistage Batch Distillation of Ethanol/Water with
Varibale Reflux ............................................................................................................................. 33
Figure 3.4 Schematic of Cyclic Distillation on Carl Still ............................................................. 38
x
Figure 3.5 Comparison of Energy Requirements in Cyclic Distillation versus Conventional
Distillation [26] ............................................................................................................................. 43
Figure 3.6 Cyclic Distillation Design Approach........................................................................... 45
Figure 4.1 Manufactured Sampling Apparatus ............................................................................. 53
Figure 5.1 Ethanol Concentration in Distillate vs Time for All Distillations ............................... 59
Figure 5.2 Ethanol Concentration vs Volume for Distillations D7 – D15 .................................... 60
Figure 5.3 Normalized Ethanol Concentration vs Volume for Distillations D7-D15 .................. 60
Figure 5.4 Normalized Acetaldehyde Concentration vs Volume for Distillations D7-D15 ......... 61
Figure 5.5 Normalized Acetone Concentration vs Volume for Distillations D7-D15 ................. 61
Figure 5.6 Normalized Ethyl Acetate Concentration vs Volume for Distillations D7-D15 ......... 62
Figure 5.7 Normalized Methanol Concentration vs Volume for Distillations D7-D15 ............... 62
Figure 5.8 Normalized Propanol Concentration vs Volume for Distillations D7-D15 ................ 63
Figure 5.9 Normalized Isobutanol Concentration vs Volume for Distillations D7-D15 .............. 63
Figure 5.10 Normalized Butanol Concentration vs Volume for Distillations D7-D14 ................ 64
Figure 5.11 Normalized Isoamyl Alcohol Concentration vs Volume for Distillations D7-D15 .. 64
Figure 5.12 Hydrometer ABV Reading vs Fraction of Volume Distilled for Low Wine
Distillations ................................................................................................................................... 65
Figure 5.13 Steam Efficiency for All Low Wine Distillations ..................................................... 66
Figure 5.14 Percent Recovery for All Low Wine Distillations .................................................... 66
Figure 5.15 Hydrometer ABV Reading vs Fraction of Volume Distilled for Finishing
Distillations ................................................................................................................................... 67
Figure 5.16 Steam Efficiency for All Finishing Distillations ....................................................... 67
Figure 5.17 Percent Recovery for All Finishing Distillations ...................................................... 68
Figure 5.18 Normalized Proof Gallon of Heads Cut by Distillation ............................................ 69
xi
Figure 5.19 Normailized Proof Gallon of Hearts Cut by Distillation ........................................... 70
Figure 5.20 Normailized Proof Gallon of Tails Cut by Distillation ............................................. 70
Figure 5.21 Temperature vs Time Profile for Conventional Operation (D11) ............................. 71
Figure 5.22 Temperature vs Time Profile for Cyclic Operation (D13) ........................................ 71
Figure 5.23 MATLAB Simulation of Conventional Distillation: Composition vs Time ............. 73
Figure 5.24 MATLAB Simulation of Conventional Distillation: Temperature vs Time ............. 74
Figure 5.25 MATLAB Simulation of Conventional Distillation: Moles in Pot vs Time ............. 75
Figure 5.26 MATLAB Simulation of Conventional Distillation: Moles in Distillate vs Time .... 76
Figure 5.27 Stage and Reboiler Composition vs Time for First Vapor Flow Period ................... 77
Figure 5.28 Stage and Reboiler Temperature vs Time for First Vapor Flow Period.................... 78
Figure 5.29 Cyclic Batch Distillation Tray and Still Composition vs Time for 25 Cycles .......... 79
Figure 5.30 Cyclic Batch Distillation Tray and Still Temperature vs Time for 25 Cycles ......... 79
Figure A.1 Normal Operation 1, Tray 1 Graphical Results .......................................................... 89
Figure A.2 Normal Operation 1, Tray 2 Graphical Results .......................................................... 89
Figure A.3 Normal Operation 1, Tray 3 Graphical Results .......................................................... 90
Figure A.4 Normal Operation 2 , Tray 1 Graphical Results ......................................................... 90
Figure A.5 Normal Operation 2 , Tray 2 Graphical Results ......................................................... 91
Figure A.6 Normal Operation 2 , Tray 3 Graphical Results ......................................................... 91
Figure A.7 Normal Operation 3 , Tray 1 Graphical Results ......................................................... 92
Figure A.8 Normal Operation 3 , Tray 2 Graphical Results ......................................................... 92
Figure A.9 Normal Operation 3 , Tray 3 Graphical Results ......................................................... 93
Figure A.10 Cyclic Distillation 1 (Vapor 9 min, Liquid 3 min) , Tray 1 Graphical Results ........ 93
Figure A.11 Cyclic Distillation 1 (Vapor 9 min, Liquid 3 min) , Tray 2 Graphical Results ........ 94
xii
Figure A.12 Cyclic Distillation 1 (Vapor 9 min, Liquid 3 min) , Tray 3 Graphical Results ........ 94
Figure A.13 Cyclic Distillation 2 (Vapor 6 min, Liquid 3 min) , Tray 1 Graphical Results ........ 95
Figure A.14 Cyclic Distillation 2 (Vapor 6 min, Liquid 3 min) , Tray 2 Graphical Results ........ 95
Figure A.15 Cyclic Distillation 2 (Vapor 6 min, Liquid 3 min) , Tray 3 Graphical Results ........ 96
Figure A.16 Cyclic Distillation 3 (Vapor 4 min, Liquid 3 min) , Tray 1 Graphical Results ........ 96
Figure A.17 Cyclic Distillation 3 (Vapor 4 min, Liquid 3 min) , Tray 2 Graphical Results ........ 97
Figure A.18 Cyclic Distillation 3 (Vapor 4 min, Liquid 3 min) , Tray 3 Graphical Results ........ 97
Figure A.19 Normal Operation – Low Wines , Tray 1 Graphical Results ................................... 98
Figure A.20 Normal Operation – Low Wines , Tray 2 Graphical Results ................................... 98
Figure A.21 Normal Operation – Low Wines , Tray 3 Graphical Results ................................... 99
Figure A.22 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 9 min, Liquid 3 min)
Tray 1 Graphical Results .............................................................................................................. 99
Figure A.23 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 9 min, Liquid 3 min)
Tray 2 Graphical Results ............................................................................................................ 100
Figure A.24 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 9 min, Liquid 3 min)
Tray 3 Graphical Results ............................................................................................................ 100
Figure A.25 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 6 min, Liquid 3 min)
Tray 1 Graphical Results ............................................................................................................ 101
Figure A.26 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 6 min, Liquid 3 min)
Tray 2 Graphical Results ............................................................................................................ 101
Figure A.27 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 6 min, Liquid 3 min)
Tray 3 Graphical Results ............................................................................................................ 102
Figure A.28 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 4 min, Liquid 3 min)
Tray 1 Graphical Results ............................................................................................................ 102
Figure A.29 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 4 min, Liquid 3 min)
Tray 2 Graphical Results ............................................................................................................ 103
xiii
Figure A.30 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 4 min, Liquid 3 min)
Tray 3 Graphical Results ............................................................................................................ 103
Figure A.31 Normal Operation – Brandy Finishing Run w/ reflux - Tray 1 Graphical Results 104
Figure A.32 Normal Operation – Brandy Finishing Run w/ reflux - Tray 2 Graphical Results 104
Figure A.33 Normal Operation – Brandy Finishing Run w/ reflux - Tray 3 Graphical Results 105
Figure A.34 Normal Operation – Brandy Finishing Run w/o reflux - Tray 1 Graphical Results
..................................................................................................................................................... 105
Figure A.35 Normal Operation – Brandy Finishing Run w/o reflux - Tray 2 Graphical Results
..................................................................................................................................................... 106
Figure A.36 Normal Operation – Brandy Finishing Run w/o reflux - Tray 3 Graphical Results
..................................................................................................................................................... 106
Figure A.37 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 1 Graphical
Results ......................................................................................................................................... 107
Figure A.38 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 2 Graphical
Results ......................................................................................................................................... 107
Figure A.39 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 3 Graphical
Results ......................................................................................................................................... 108
Figure A.40 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 1 Graphical
Results ......................................................................................................................................... 108
Figure A.41 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 2 Graphical
Results ......................................................................................................................................... 109
Figure A.42 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 3 Graphical
Results ......................................................................................................................................... 109
Figure A.43 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 1 Graphical
Results ......................................................................................................................................... 110
Figure A.44 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 2 Graphical
Results ......................................................................................................................................... 110
Figure A.45 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 3 Graphical
Results ......................................................................................................................................... 111
xiv
Figure A.46 Normal Operation 1 – Distillate Graphical Results ................................................ 112
Figure A.47 Normal Operation 2 – Distillate Graphical Results ................................................ 112
Figure A.48 Normal Operation 3 – Distillate Graphical Results ................................................ 113
Figure A.49 Cyclic Distillation 1 (Vapor 9 min, Liquid 3 min) – Distillate Graphical Results 113
Figure A.50 Cyclic Distillation 1 (Vapor 6 min, Liquid 3 min) – Distillate Graphical Results 114
Figure A.51 Cyclic Distillation 1 (Vapor 4 min, Liquid 3 min) – Distillate Graphical Results 114
Figure A.52 Normal Operation – Low Wines – Distillate Graphical Results ............................ 115
Figure A.53 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 9 min, Liquid 3 min)
Distillate Graphical Results ........................................................................................................ 115
Figure A.54 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 6 min, Liquid 3 min)
Distillate Graphical Results ........................................................................................................ 116
Figure A.55 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 4 min, Liquid 3 min)
Distillate Graphical Results ........................................................................................................ 116
Figure A.56 Normal Operation – Brandy Finishing Run w/o reflux - Distillate Graphical Results
..................................................................................................................................................... 117
Figure A.57 Normal Operation – Brandy Finishing Run w/ reflux - Distillate Graphical Results
..................................................................................................................................................... 117
Figure A.58 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Distillate Graphical
Results ......................................................................................................................................... 118
Figure A.59 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Distillate Graphical
Results ......................................................................................................................................... 118
Figure A.60 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Distillate Graphical
Results ......................................................................................................................................... 119
Figure A.61 Normal Operation 1 – Bottom Column Graphical Results..................................... 120
Figure A.62 Normal Operation 2 – Bottom Column Graphical Results..................................... 120
Figure A.63 Normal Operation 3 – Bottom Column Graphical Results..................................... 121
xv
Figure A.64 Cyclic Distillation 1 (Vapor 9 min, Liquid 3 min) – Cycled Volume Graphical
Results ......................................................................................................................................... 121
Figure A.65 Cyclic Distillation 1 (Vapor 6 min, Liquid 3 min) – Cycled Volume Graphical
Results ......................................................................................................................................... 122
Figure A.66 Cyclic Distillation 1 (Vapor 4 min, Liquid 3 min) – Cycled Volume Graphical
Results ......................................................................................................................................... 122
Figure A.67 Normal Operation – Low Wines – Bottom Column Graphical Results ................. 123
Figure A.68 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 9 min, Liquid 3 min)
Cycled Volume Graphical Results .............................................................................................. 123
Figure A.69 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 6 min, Liquid 3 min)
Cycled Volume Graphical Results .............................................................................................. 124
Figure A.70 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 4 min, Liquid 3 min)
Cycled Volume Graphical Results .............................................................................................. 124
Figure A.71 Normal Operation – Brandy Finishing Run w/o reflux - Bottom Column Graphical
Results ......................................................................................................................................... 125
Figure A.72 Normal Operation – Brandy Finishing Run w/ reflux - Bottom Column Graphical
Results ......................................................................................................................................... 125
Figure A.73 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Cycled Volume
Graphical Results ........................................................................................................................ 126
Figure A.74 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Cycled Volume
Graphical Results ........................................................................................................................ 126
Figure A.75 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Cycled Volume
Graphical Results ........................................................................................................................ 127
Figure A.76 Acetaldehyde Concentration vs Time for All Distillations .................................... 128
Figure A.77 Acetone Concentration vs Time for All Distillations ............................................. 128
Figure A.78 Ethyl Acetate Concentration vs Time for All Distillations .................................... 129
Figure A.79 Methanol Concentration vs Time for All Distillations ........................................... 129
Figure A.80 Ethanol Concentration vs Time for All Distillations .............................................. 130
xvi
Figure A.81 Propanol Concentration vs Time for All Distillations ............................................ 130
Figure A.82 Isobutanol Concentration vs Time for All Distillations ......................................... 131
Figure A.83 Butanol Concentration vs Time for All Distillations.............................................. 131
Figure A.84 Isoamyl Alcohol Concentration vs Time for All Distillations ............................... 132
Figure C1. Example Chromatograph Results ............................................................................. 142
xvii
KEY TO SYMBOLS
Nomenclature
𝑥 − 𝑙𝑖𝑞𝑢𝑖𝑑 𝑚𝑜𝑙𝑒 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛
𝑦 − 𝑣𝑎𝑝𝑜𝑟 𝑚𝑜𝑙𝑒 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛
𝑘 𝑜𝑟 𝐾 − 𝑒𝑞𝑢𝑖𝑙𝑏𝑟𝑖𝑢𝑚 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡
𝐹 − 𝑓𝑒𝑒𝑑
𝑊 − ℎ𝑜𝑙𝑑 𝑢𝑝 (𝑙𝑖𝑞𝑢𝑖𝑑)
𝑈 − ℎ𝑜𝑙𝑑 𝑢𝑝 (𝑣𝑎𝑝𝑜𝑟)
𝐷 − 𝑑𝑖𝑠𝑡𝑖𝑙𝑙𝑎𝑡𝑒
𝐵 − 𝑏𝑜𝑡𝑡𝑜𝑚𝑠
𝑀 − ℎ𝑜𝑙𝑑𝑢𝑝
𝑉 − 𝑣𝑎𝑝𝑜𝑟 𝑓𝑙𝑜𝑤 𝑟𝑎𝑡𝑒
𝐿 − 𝑙𝑖𝑞𝑢𝑖𝑑 𝑓𝑙𝑜𝑤 𝑟𝑎𝑡𝑒
𝑅 − 𝑟𝑒𝑓𝑙𝑢𝑥
𝑄 − ℎ𝑒𝑎𝑡 𝑑𝑢𝑡𝑦
𝐻 𝑜𝑟 ℎ − 𝑒𝑛𝑡ℎ𝑎𝑙𝑝𝑦/𝑠𝑝𝑒𝑐𝑖𝑓𝑖𝑐 𝑒𝑛𝑡ℎ𝑎𝑙𝑝𝑦
𝛾 − 𝑎𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡
𝑡 − 𝑡𝑖𝑚𝑒
𝑃 − 𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒
𝑇 − 𝑡𝑒𝑚𝑝𝑒𝑟𝑎𝑡𝑢𝑟𝑒
𝑁𝑇 − 𝑡𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠𝑡𝑎𝑔𝑒𝑠
𝑂𝐿 − 𝑞𝑢𝑎𝑛𝑡𝑖𝑡𝑦 𝑜𝑓 𝑙𝑖𝑞𝑢𝑖𝑑 𝑡ℎ𝑎𝑡 𝑓𝑙𝑜𝑤𝑠 𝑓𝑟𝑜𝑚 𝑠𝑡𝑎𝑔𝑒 𝑡𝑜 𝑠𝑡𝑎𝑔𝑒 𝑑𝑢𝑟𝑖𝑛𝑔 𝑡ℎ𝑒 𝑙𝑖𝑞𝑢𝑖𝑑 𝑓𝑙𝑜𝑤 𝑝𝑒𝑟𝑖𝑜𝑑
xviii
Super Scripts
(𝑉) − 𝑒𝑛𝑑 𝑜𝑓 𝑣𝑎𝑝𝑜𝑟 𝑓𝑙𝑜𝑤 𝑝𝑒𝑟𝑖𝑜𝑑
(𝐿) − 𝑒𝑛𝑑 𝑜𝑓 𝑙𝑖𝑞𝑢𝑖𝑑 𝑓𝑙𝑜𝑤 𝑝𝑒𝑟𝑖𝑜𝑑
𝑐 − 𝑐𝑦𝑐𝑙𝑒 𝑛𝑢𝑚𝑏𝑒𝑟
Subscripts
𝑖, 𝑗, 𝑘 − 𝑠𝑡𝑎𝑔𝑒 𝑛𝑢𝑚𝑏𝑒𝑟
𝐹 − 𝑓𝑒𝑒𝑑
𝐷 − 𝑑𝑖𝑠𝑡𝑖𝑙𝑙𝑎𝑡𝑒
𝐵 − 𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 𝑜𝑟 𝑏𝑜𝑡𝑡𝑜𝑚𝑠
𝑠𝑎𝑡 − 𝑠𝑎𝑡𝑢𝑟𝑎𝑡𝑖𝑜𝑛
𝑣𝑎𝑝 − 𝑣𝑎𝑝𝑜𝑟 𝑓𝑙𝑜𝑤
1
CHAPTER 1: Introduction
1.1 Distillation Theory and Background
Distillation technology has been used worldwide throughout the chemical industry for
many years, originally being invented in China around 800 BC. There are currently over 40,000
columns in operation worldwide [29]. Although there are many variations in column types, sizes,
tray configurations, etc., the main principle is the same. Distillation is the physical process of
heating a liquid mixture to separate components based on the relative volatilities of each
component.
There are primarily two main types of distillation, continuous distillation and batch
distillation. Continuous distillation requires a continuous feed into the column at an optimal tray,
bottoms product and distillate product are continuously collected. This process can theoretically
continue forever at steady state if there is feed to the column. Multiple feeds can be fed at different
locations as well as multiple product distillates may be taken off at different stages in the column.
In batch distillation a still pot is used to hold a certain initial volume of liquid to be separated,
known as the feed. The mixture is heated, and distillate is collected until processing of the feed is
completed. The distillate may be collected all together or ‘cuts’ maybe be taken as concentrations
in the distillate change over time. In both types of distillation, heat is supplied by direct fire, steam,
or electric heating. As the mixture is heated, the most volatile component, the one with the lowest
boiling point, will vaporize first and travel upward through the column. The column can have any
number of trays ranging from 3 up to 40 or more depending on the size of production and products
being made. Trays allow for rectification of the feed throughout the column. There are many
different tray types, but each serve the purpose of allowing some of the vapor to condense to liquid
2
on the tray and some of the vapor to continue up the column. Each additional tray increases
rectification and causes the vapor at the top of the column to have the highest concentration of the
most volatile component at any time during the distillation. Finally, the vapor may travel through
either a partial or total condenser. Partial condensers partially condense the vapor sending some
liquid back into the column, known as reflux. Total condensers totally condense the vapor into
liquid and the product exits the system, known as the distillate. In batch distillation there are two
operation modes that are commonly discussed in academia, constant reflux or constant distillate
composition. When the reflux is held constant the distillate will initially start with a high
concentration of the most volatile component in the system and gradually decrease over time. If a
constant distillate composition is desired, then the reflux ratio can be adjusted throughout the
distillation. In general, increasing the reflux increases rectification, so as the distillate composition
begins to decrease, the reflux will be increased to keep the distillate at a constant composition.
1.2 Distillation Improvement and Cyclic Distillation
Distillation is the most widely used separation method. Consequently, distilleries
consume about 40% of the total energy used to operate plants in the chemical industry, and over
95% of the energy used in separation processes [31]. Therefore, improving distillation
technology is one area that chemical engineers are working to transform into a more efficient
process by use of process intensification. Process intensification (PI) is a design philosophy that
aims to improve the efficiency of an already existing machine or process. In general, it can be a
change in equipment or a change in a method to increase efficiency. In distillation, PI aims to
improve product quality, lower energy requirements, increase capacity, reduce time from raw
3
material to market and increase safety by design. Overall, PI aims to safely increase the speed
and quality of production while decreasing capital and operating costs. The four domains that PI
should make use of include: spatial (structure), thermodynamic (energy), functional (synergy), and
temporal (time) [30]. There are many PI distillation technologies that have been suggested
including heat pump assisted distillation, membrane distillation with its various designs, HiGee
technology, cyclic distillation, dividing wall column, and reactive distillation [29].
The purpose of this research was to investigate the possible economic and energy impacts
of cyclic distillation as well as its application to the distilled spirits industry. Cyclic distillation is
an alternate mode of operation first proposed and studied by Cannon and McWhirter in the 1960’s.
The ‘cycle’ consists of alternating between two periods, vapor flow and liquid flow. During the
vapor flow period heat is supplied to the column, vapor flow upwards and distillate is collected as
in normal operation. During the liquid flow period, the heat source is ceased and the liquid on each
tray is transferred to the tray below. The cycle of alternating between the two periods is continued
for the entire distillation. Cyclic distillation has been applied to many different systems and
configurations and has been demonstrated experimentally and theoretically to increase column
throughput, lower energy requirements and achieve higher separation performance. Previous
studies on cyclic distillation and current usage will be discussed in the proceeding chapter.
1.3 Cyclic Distillation Application to Spirits Industry
The distilled spirits industry is a rapidly growing industry. The US Distilled Spirits Council
reported an 8th consecutive year of market share gains in 2017, “supplied sales were up 4 percent,
rising $1 billion to a total of $26.2 billion, while volumes rose 2.6 percent to 226 million cases, up
5.8 million cases from the prior year. These results reflect adult consumers’ ongoing taste for
4
higher-end distilled products across most categories [40].” Increasing demand for higher-end
products causes manufactures to look for technologies that can increase efficiency of their process
and profit for their company. Cyclic distillation can help meet and exceed these higher demands.
Distilled spirits are different from other distilled products in that they are eventually going
to be consumed as a beverage by the consumer. In many processes the end goal is to concentrate
one or more products with as high purity as possible, in distilled spirits this is not necessarily the
case. To make a distilled spirit product, first the grain is milled into a flour like constituency. The
grain is then mixed with water to form a slurry, known as the mash, in a large vat called a mashtun.
During the mashing process, the mash is heated, and alpha and beta amylase enzymes are added
at the appropriate time, temperature and pH to enable the breakdown of starch into simple sugars,
like glucose, that yeast can consume. The mash is then cooled and transferred to a fermenter to
ferment for 1-2 weeks. During fermentation of the mash, yeast produce mainly ethanol and carbon
dioxide but also by products including other alcohols as well as esters, aldehydes, and ketones.
The distilling industry refers to the other compounds as congeners. The concentration of the
fermentation at the end is roughly 8-12% alcohol by volume. The next step in the process is
stripping the mash into low wines. The fermented mash is pumped into a designated stripping still
which may or may not have trays. The purpose of stripping is to separate and concentrate the
volatile components formed during fermentation from the solid and non-volatile components. The
distillate called low wines is collected entirely and is usually 25-40% alcohol by volume. The next
step is the finishing run in another distillation column. In large operations, this is done on a
continuous column and the product is pulled off the column at the appropriate ethanol
concentration for the product being made.
5
Batch distillation for both the stripping and finishing run is used widely throughout the
craft spirits industry. Many craft distilleries pride themselves on their beautiful copper stills that
look like pieces of art in their distilleries. In batch finishing, a certain volume of low wines is added
to the still and distillate is collected. During batch distillation of spirits, fractions of the distillate
are separated into different volumes known as cuts. The first distillate product is known as the
heads cut, which is heavily concentrated with ethanol as well as the most volatile components
mainly acetone, acetaldehyde, methanol and ethyl acetate. Once those compounds are out of the
system, detected by the distiller by the flavor and aroma of the distillate, the actual product is
collected and known as hearts. The hearts cut has primarily ethanol, but other compounds also
come through in low concentrations such as methanol, isoamyl alcohol and other higher alcohols.
Towards the end of the distillation the higher alcohols start to increase in concentration in the
distillate and the flavor and aroma of the distillate become undesirable; this is when the tails cut is
made. The tails cut still contains a high amount of ethanol, so it is oftentimes re-distilled to collect
as much hearts as possible. Since the ethanol concentration is high and has other hazardous
compounds, it cannot go down the drain and must be disposed of properly. Tails storage and
disposal is a big problem from smaller distilleries with small facilities while trying to balance the
energy requirement of re-distillation with the actual profit it lends. While ethanol is the primary
contributor to the average spirit flavor and aroma profile, concentrations near the ppm level of the
other compounds mentioned are found and desired to make a balanced and flavorful product.
Therefore, it is important to investigate the trends of the congeners during cyclic distillation. Thus
far cyclic distillation has not been applied to distilling spirits. The next few chapters intend to
investigate a thorough literature review of cyclic distillation, to model the proposed batch cyclic
6
distillation method, as well as to present the experimental methods and data collected while
cyclically distilling apple brandy on an industry sized batch column still.
7
CHAPTER 2: Literature Review
Cyclic distillation was first proposed by Cannon in 1961. The ‘cycle’ he proposed
consisted of two periods, a vapor flow period, and a liquid flow period. Cannon used a cycle
timer with an automated valve to allow the periods to be specifically timed. While the valve was
open, the vapor flowed through the column, and did not allow liquid to travel down through the
trays. When the valve was closed, the lack of vapor thrust allowed the liquid to fall to the preceding
tray. This removed the need for downcomers on plates because the liquid and vapor flowed through
the same area at different times. Cannon concluded the advantages of controlled cycling in
distillation were higher capacity, simpler and cheaper plate design, and high flexibility due to a
choice of operating conditions dependent on cycle times [1].
Cannon and others applied controlled cycling to sieve and screen plate towers as well as
packed-plate columns [2,3]. Gaska and Cannon reported experimental data on a distillation of a
mixture of benzene and toluene using two test towers, one with 9 plates spaced 18 ½ inches
apart, and the other with 17 plates spaced 9 ½ inches apart. The data showed a 48% increase in
total vapor load at a fixed column pressure drop. They reported a significant increase in column
capacity based on the average vapor velocity achieved which is shown graphically below.
Figure 2.1 Cycling Increases Column Capacity for Type-Two Plates [2]
8
Gaska and Cannon concluded that the maximum rate of phase flow was not dictated by solely the
equipment size and properties of the system, but was also dependent on the mode of operation of
the column [2]. McWhirter and Cannon also found that it was possible to produce maximum tower
efficiency, maximum tower capacity or anywhere in between these via controlled cycling of a
packed plate tower [3]. McWhirter later demonstrated in his Ph.D. thesis that it was possible to
double the stage efficiency and triple the throughput of a column simultaneously through
theoretical and experimental investigations. McWhirter also demonstrated that while over multiple
cycles the column appeared to be in steady-state operation, during a single cycle it was in unsteady-
state. At any instance during the vapor flow period, the composition of the liquid on a tray varied
with time, and thus the composition of the vapor leaving the stage also varied with time. He also
showed that the average driving forces for mass transfer in a cyclic column were considerably
greater than those in a conventional column. McWhirter believed that the high stage efficiencies
of a cyclic column were difficult, if not impossible, to rationalize in terms of conventional
operation. Lastly, McWhirter used these ideas to develop computer simulations of cycled columns
by solving the unsteady-state material balance equations describing compositions as a function of
time by using the finite-difference method [3]. Schrodt was the next to study controlled cyclic
distillation and implement this operation mode on a plant-scale [4,5,6,7]. First Schrodt and others
re-introduced a simple model to describe the controlled cycle column [6]. They constructed figures
relating vapor compositions 𝑥𝑖𝑉 as a function of the fraction of plate holdup dropped (𝜙).
9
Figure 2.2 Compositions 𝒙𝒊 𝑽 as a Function of Fraction of Plate Holdup Dropped (𝝓) [6]
Figure 2.2 illustrates the concept that with the closer to true plug flow and more cycles, the purer
the resulting vapor composition, which is not dependent on the number of plates. Schrodt defined
the effective plate efficiency as:
𝐸0,𝑖 =𝑥𝑖+1
𝑉 − 𝑥𝑖𝑉
𝑘𝑖𝑥𝑖𝑉 − 𝑥𝑖
𝑉(2.1)
Where, 𝑘𝑖, is a term used to calculate the equilibrium relationship. In addition, Figure 2.3 was
generated which illustrated the effective plate efficiency as a function of plate number.
10
Figure 2.3 Effective Plate Efficiency 𝑬𝟎 as a Function of Plate Number [6]
Next, Schrodt and others developed a computational method for obtaining the theoretical
number of stages similar to the McCabe-Thiele. The current method will be presented later in this
paper. The next paper published by Schrodt and others reported the results of computer
simulations of controlled cyclic distillation to investigate the theoretical effects of various
parameters on the separating ability of controlled cycling [5]. At this point is important to remind
the reader of the definition of various efficiencies below in Table 2.1.
Table 2.1 Definition of Various Efficiencies [5]
Efficiency Symbol Definition
Murphree (vapor) point
efficiency 𝐸 Local vapor efficiency at a
point in time and space
Instantaneous plate efficiency -
Over-all vapor efficiency of a
plate at a point in time
Effective plate efficiency 𝐸0 Over-all efficiency of a plate
computed on the basis of
liquid-phase plate
compositions
Over-all column efficiency 𝐸0 Number of theoretical stages
(𝐸0 = 100%) corresponding
to a given separation divided
by the number of actual plates
in the column
11
The Murphree point efficiency is assumed to be the same and equal for all plates and is defined
as:
𝐸 =𝑦𝑛 − 𝑦𝑛−1
𝑦𝑛∗ − 𝑦𝑛−1
(2.2)
Where 𝑦𝑛 ∗ is the composition of the vapor that would be in equilibrium with the liquid on the nth
stage and is given by the vapor-liquid equilibrium relationship assumed.
For computer simulations, Schrodt used the previously developed equations to see how the
plate compositions varied with time as well as how the column approached its pseudo-steady-state
conditions [5]. The pseudo steady state was defined as the composition profile remaining constant
at the end of successive cycles. For the simulation, a binary system with a constant volatility of
1.2 was assumed. It ran in total reflux and assumed a boil-up rate and liquid hold up of 0.1 g
mole/sec and 0.2 g mole, respectively. In addition, the initial pot was charged with 10 g moles, the
Murphree efficiency was assumed to be 1, the fraction of plate holdup dumped to the plate below
was assumed to be 1, and N was assumed to be 5. Lastly, the initial composition of the still pot
was 0.5.
The results from the computer simulation gave graphical insight to how a controlled cycled
column operates. It was found that the pseudo steady state condition was achieved after roughly
150 cycles.
The effect of fraction of plate holdup dropped was studied for values of 𝜙 ranging from 0
to 3.0. The effective numbers of theoretical plates in the column (𝑁𝑒𝑓𝑓) was calculated from the
values of the still-pot and condenser compositions at the end of the vapor flow period once the
pseudo-steady-state condition was achieved, using the Fenske equation below:
12
𝑁𝑒𝑓𝑓 = (log [
𝑥𝑐
1 − 𝑥𝑐 1 − 𝑥1
𝑥1]
log 𝛼) − 1 (2.3)
The overall column efficiency (𝐸0) was then determined by dividing 𝑁𝑒𝑓𝑓 by the number of
actual plates. The resulting data can be seen below in Figure 2.4.
Figure 2.4 Effect of Fraction of Plate Holdup Dropped During the Liquid Flow Period on
the Over-all Efficiency of a Controlled Cycling Rectification Still at Total Reflux [5]
The results from Schrodt’s computer simulations agreed with McWhirter’s theoretical
studies that maxima occur at integral values of 𝜙, with the maximum of these occurring at 𝜙 = 1.
It is important to note that 𝜙 = 0 corresponds to conventional column operation.
The next parameter studied was the effect of mixing during the liquid flow period. Values
of 𝜏 equal to 0.5, 1.0 and 2.0 seconds were investigated which correspond to 𝜙 = 0.25, 0.5 and 1.0,
respectively. Results of these simulations can be seen in Table 1.2 below.
13
Table 2.2 Effects of Mixing During the Liquid Flow Period on the Separating Ability of a
Controlled Cycle Rectification Still at Total Reflux [5]
Duration of the vapor-flow
period 𝜏(= 𝜏𝐿), 𝑠𝑒𝑐
Effective number of
theoretical plates, 𝑁𝑒𝑓𝑓
Over-all column efficiency,
𝐸0, %
0.5 4.96 99.1
1.0 4.93 98.7
2.0 4.83 96.6
The data presented in Table 2.2 illustrates that complete mixing during the liquid flow period
decreased the separation advantages that were gained in cyclic operation.
The next effect studied was the effect of relative volatility. The results are shown below in
Table 2.3.
Table 2.3 Effect of Relative Volatility on the Overall Column Efficiency of a Controlled
Cycling Rectification Still at Total Reflux [5]
Relative volatility, 𝛼 Effective number of
theoretical plates, 𝑁𝑒𝑓𝑓
Over-all column efficiency,
𝐸0, %
1.05 9.56 191.1
1.10 9.54 190.8
1.20 9.46 189.2
1.30 9.35 187.0
1.40 9.23 184.6
1.50 9.12 182.4
1.50a 9.26 185.3 a 𝑥1
0 = 0.3
The results in Table 2.3 show that the overall column efficiency decreased with an increase in the
relative volatility of the distilled mixture.
Next, the effect of Murphree plate efficiency was studied and it was concluded that the
over-all column efficiency of a controlled cyclic distillation column increased rapidly as the
individual plate efficiencies were increased. This conclusion was also in agreement with
McWhirter’s earlier findings. Lastly, the effect of the number of plates, N, was studied via
computer simulations. In this case, a straight-line equilibrium relationship was assumed. From
14
these simulations, it was found that the over-all column efficiency of a controlled cycling column
increased to an asymptotic value as the number of actual plates in the still increased. These findings
are illustrated in Figure 2.5 below.
Figure 2.5 Effective Plate Efficiencies (E0) in a Controlled Cycling Rectification Column
Still at Total Reflux [5]
To study plant-scale operation, a 20-plate column was designed to separate a mixture of
acetone and water [7]. Schrodt and others found that the use of a conventional column in a cyclic
operational mode resulted in substantial capacity improvements, if the number of trays was less
than 12. This was assumed to be due to pressure drops throughout the column which caused trays
to mix with each other during the liquid flow period. They suggested that an 100% efficiency
increase, and 2 to 3 times capacity increases were still possible if true plug flow was observed.
15
True plug flow would be achieved if the volume from tray 4 only dropped to tray 3, and none of
the liquid from tray 5 would mix with the liquid of tray 3 [7].
In 1967, Robinson and Engel presented a theoretical analysis that demonstrated the
advantages of cycled mass transfer operations [9]. Considering a classic paper by Lewis on the
effect of liquid phase composition profiles applicable to cyclic operation, Robinson and Engel
developed time-axis concentration gradients on each stage, analogous to those developed for
conventional columns with lateral concentration gradients to predict column performance. Lewis’
paper considering three cases. Case 1, vapor flowing to the tray, liquid flows at uniform
composition at all points. Case 2, vapor laterally unmixed flowing between stages, liquid flows in
same lateral direction on all stages. Case 3, vapor laterally unmixed flowing between stages, liquid
flows in opposite directions on alternate stages. Robinson and Engel decided that the plate directly
above the reboiler was analogous to a Lewis Case 1 with constant composition vapor flowing to it
over short times of one cycle. The rest of the plates corresponded to Lewis Case 2 plates. Robinson
and Engel then used the mathematical analyses developed by Lewis to derive a vapor flow period
material balance and solutions to case 1 and case 2. The plate efficiency was found to be dependent
on the fraction of liquid on the plate that drained to the plate below during cycling. The results of
this analysis were like the results of McWhirter and Schrodt, therefore the equations and figures
have not been presented [9].
In 1967, Horn published a paper on periodic countercurrent processes. The report discussed
how the overall stage efficiency of a periodically operated distillation column is complicated and
depends on the number of stages as well as the equilibrium and transport parameters. Horn
developed accurate asymptotic formulas for the stage efficiency that proved the idea that the
performance of a distillation column can be drastically improved by periodic operation [8].
16
In 1968, Horn and May developed a simple asymptotic relation which described the stage
efficiency of a periodically operated distillation column. The relation developed used parameters
which are a function of Murphree efficiency, transport number, and separation factor. Horn and
May defined an important number, the transport number as the ratio of the quantity of liquid
transported per cycle to the liquid hold up of a stage. They also investigated the effect of mixing
on stage efficiency of a periodically operated countercurrent process for the case of difficult
separations. They did so by modeling each stage, during the time of liquid transport, by a series of
well stirred tanks [10,11].
Gel’perin, Polotskii, and Potapov experimented with a bubble-cap fractionating column in
a cyclic regime [13]. The column contained six sections having an internal diameter of 80 mm and
a height of 250 mm, each with five single cap plates in between the sections. The experiment
allowed for varying total cycle time from 5 to 60 seconds, operated at total reflux and atmospheric
pressure. To incorporate the cyclic regime, electromagnetic solenoid valves were used in the vapor
feed and reflux lines and controlled by two-time relays through the rectifier. From the resulting
data, they determined an optimum total cycle time and ratio of vapor to liquid feed periods. The
experiments confirmed the presence of a beneficial effect from the use of the cycle regime with
bubble-cap plates as seen in Figure 1.5 below.
17
(a) (b) (c)
Figure 2.6 (a) Dependence of Column Separation Efficiency E on Mean Vapor Velocity 𝑾𝒗
(b) Dependence of Column Separation Efficiency E on Vapor Feed Time 𝝉𝒗 and on Liquid-
feed Time 𝝉𝒍 (c) Dependence of Column Separation Efficiency E on the Ratio of Feed
Periods of Phases 𝝉𝒗/𝝉𝒍 for Various Velocities [13]
In the Figure 1.6(a) line 1 is the stationary regime, line 2 is cyclic regime with 𝜏𝑣 = 8 𝑠𝑒𝑐 and
𝜏𝑙 = 5 𝑠𝑒𝑐, and line 3 is cyclic regime with 𝜏𝑣 = 12 𝑠𝑒𝑐 and 𝜏𝑙 = 3 𝑠𝑒𝑐. The cyclic regime was
clearly more efficient than the stationary mode of operation. The maxima in the cyclic regime lines
indicates a maximum vapor flow rate corresponding to a maximum efficiency. In Figure 1.6(b)
line 1 is for 𝜏𝑣 = 8 𝑠𝑒𝑐, line 2 is for 𝜏𝑣 = 12 𝑠𝑒𝑐, line 3 if for 𝜏𝑙 = 5 𝑠𝑒𝑐 and line 4 is for 𝜏𝑙 =
1 𝑠𝑒𝑐. This figure allowed for the assumption of the optimal vapor feed time of 8 seconds and the
optimal liquid feed time to be 5 seconds. It is important to note that the optimal duration of liquid
feed to the column was determined by the time to replace the liquid on each plate and was a
function of plate construction and properties of the liquid. In Figure 1.6(c) line 1 corresponds to a
mean vapor velocity 𝑊𝑣 = 1.0 𝑚/𝑠𝑒𝑐 and line 2 corresponds to 𝑊𝑣 = 0.8 𝑚/𝑠𝑒𝑐. From the data
it is evident that for this column, the optimum total cycle time is about 13 sec, and the ratio of the
feed periods for vapor and liquid is in the range from 1 to 3 [13].
18
Up to this point, most of the research done was comparing cyclic columns to conventional
operation. In 1977, Rivas developed a short cut method for preliminary design of ideal cyclic
distillation columns. Material balance differential equations on each plate of a cyclic distillation
column were used to develop analytical equations to calculate the ideal number of trays for cyclic
countercurrent processes like distillation, absorption, and stripping [14].
One year later, Furzer studied fluid flow in a distillation column by taking measurements
of the discrete residence time distribution [16]. The column had five sieve plates spaced 635 mm
apart. A microprocessor was used to obtain periodic control. The data yielded the parameters in
the (2S) model which described the fluid flow. Furzer concluded that in order to achieve maximum
separation, modifications to the column internals were required. A few years later in 1980, Furzer
and Goss studied fluid mixing and mass transfer separations of mixtures of methylcyclohexane
and n-heptane under periodically cycled conditions [17]. The theoretical improvements of 200%
reported by McWhirter and Lloyd could not be reproduced. Goss and Furzer concluded that liquid
bypassing the plates caused the reduction to 107-126% improvement. However, the (2S) model
was successful in modeling both the fluid mixing and the mass transfer separations in the
periodically cycled column.
Baron, Wajc and Lavie developed a theory of stepwise periodic distillation [18,19]. The
major difference in stepwise periodic versus controlled cycling is in stepwise the liquid flow is
controlled directly instead of through pulsations of the vapor flow rate. The researchers developed
this based on the idea that preventing axial mixing could improve the separation. Their research
goal was to compare the two operating models for total reflux and for continuous distillation, as
well as construct a lab-scale stepwise periodic column for data collection. The column analyzed
19
consisted of a boiler, a total condenser, N trays numbered starting at the bottom and N reservoirs
specific to each tray. The schematic can be seen below in Figure 2.7.
Figure 2.7 Schematic of Column Used for Stepwise Period Distillation [18]
Each tray is fitted with an inlet and an outlet with on-off valves. The trays had no
downcomers but could be completely emptied via the outlet to the reboiler. The reservoirs
collected the condensate from each tray via the inlet. One periodic cycle started with vapor alone
flowing through the column exchanging mass with the stationary liquid on each tray. The
condensate was then collected in the reservoirs corresponding to each tray. When the bottom tray
(tray N) was full, all outlet valves were opened to empty all plates into the boiler. Next, the outlet
values were closed, and inlet valves were opened to all the reservoirs to empty to their
corresponding plates. Lastly, all valves were closed again, and the cycle starts again. Baron and
others compared controlled cycling and stepwise periodic distillation. They showed that the two
processes have the same asymptotic efficiencies for large values of N while periodic distillation is
20
10 to 30% more efficient for a finite number of trays [18]. In a follow-up study, they described
separating a binary mixture using stepwise periodic operation mode. They proposed a model and
simulation algorithm for both controlled and periodic cycling. Through an extensive parametric
study, it was concluded again that stepwise periodic operation was superior to ideal controlled
cycling [19].
In 1984, Matsubara, Watanbe, and Kurimoto analyzed the work of Cannon and Baron-
Wajc in hopes to combine the advantages of both schemes [20]. They successfully developed a
composite scheme which could vary from the ideal cannon scheme to the Baron-Wajc scheme by
changing various parameters. In a second paper published in 1985, a laboratory-scale periodic
distillation column using the idealized Cannon scheme was constructed and experimented with to
view energy conservation performance [21]. They used a five-stage column to separate water and
methanol. The experimental results showed an 84.6% increase in energy efficiency due to an
average vapor flow rate roughly 50-20% lower than in conventional continuous column operation
and achieving the same separation.
Around the same time, Thompson and Furzer developed hydrodynamic modeling for liquid
holdup in periodically cycled plate columns [22]. Their objective was to predict the holdup
distribution in a periodically-cycled column, for a specific set of design and operating conditions.
They confirmed their models were satisfactory by experimenting with a 600 mm diameter Perspex
column containing four sieve plates. The column was fitted with Time Delay Plates to improve
plug flow pressure equalization at the start of the liquid flow period [22]. Szonyi and Furzer then
developed a new tray design to increase column performance using a periodic cycle operating
method [23]. They ran computer simulations which predicted 200% greater column performance
for systems with nonlinear equilibrium curves. They also conducted experiments distilling
21
methanol-water mixtures to verify the computer predictions. Using a single-plate the simulations
expectations were confirmed. A new tray design was then implemented in a five-plate insulated
mild steel column of 610mm I.D. The new trays consisted of a sieve tray plus inclined surfaces
and a vessel in between each sieve tray. The new tray design can be seen in Figure 2.8 below.
Figure 2.8 New Tray Design for Period Cycling Distillation [23]
The trays collected liquid from the sieve plate above and directed it via the inclined surface to the
vessel. In doing so, the liquid in the vessel experienced a time delay before flowing to the sieve
plate below. The trays were experimentally effective at obtaining true plug flow but were limited
by lack of ventilation and consequently were only able to achieve 140% greater column
efficiencies compared to conventional distillation. Szonyi and Furzer recommend further
improvements be made by hardware modification [23].
22
In 1989, Toftegard and Jorgensen developed an integration method for the dynamic
simulation of cycled processes. Application of the method was demonstrated on a simulation of
controlled cycling distillation. The time to compute the transient was reduced by 90% compared
to using a conventional integration method [37].
Finally, in 1997 Sorensen and Prenzler applied cyclic distillation theory to batch distillation
[38]. The cycle applied to a batch column was a repeated filling and dumping of the reflux drum.
In the beginning of the cycle the reflux drum was filled. Then the column ran under total reflux,
and the maximum attainable separation in the column was achieved. During total reflux the light
component accumulated in the reflux drum until the column reached equilibrium or steady state.
At the end of the cycle, the drum was dumped, and the product was withdrawn. An illustration of
each period can be seen below in Figure 2.9.
Figure 2.9 The Three Characteristic Periods in the Cyclic Operation of a Regular Batch
Distillation Column [39]
Sorensen concluded that the best operating procedure was to adjust the reflux drum holdup online
based on measurements of distillate composition as a function of time. If the purity is too low the
23
drum holdup could be reduced and vice versa. A laboratory batch column with 8 sieve trays was
used to implement this method [39]. A schematic of the column can be seen below in Figure 2.10.
Figure 2.10 Lab Batch Column Used to Apply a Batch Cyclic Operating Theory [39]
The still consisted of a 12 L reboiler, a column with 8 sieve trays, inner diameter of 50 mm, a
condenser, a reflux drum with a maximum holdup of 4 L, a reflux valve, and a top product receiver.
The reflux valve was a three-way solenoid valve. Samples could be taken through a manual valve
as the condensate was returned through the column. In addition, samples of the reboiler and top
product were taken. The feed consisted of 10 L of a 25 mol% methanol and 75 mol% water. The
24
required amount of product was 4 L with recovery of methanol at least 75 vol%, which
corresponded to 65 mol%. The column started with a cold feed which was heated to the boiling
point at roughly 78 ⁰C. The first cycle started when the mixture began to boil, the reflux drum
began to fill, and temperatures in the column increased. Once the drum was filled to a desired
level, reflux back to the column started, and the temperatures in the column started to decrease.
This was the beginning of the second total reflux period. Once the temperatures in the column
stabilized, steady state was achieved which happened after about 95 minutes of operation. The
reflux drum was then dumped to the product receiver for a duration of 1 min. A new cycle was
then started, and the previous procedure was repeated for another two cycles. At the end of 305
minutes of operation, 68mol% methanol was achieved in the final product which was within the
specifications. The policy was found to be easy to implement with the only interaction being the
dumping of the reflux drum. The main difficulty was determining the end of the total reflux period
which was chosen to be when the temperatures in the column did not change more than 0.1 ⁰C
within in a 5 min period [39].
In 2000, Bausa and Tsatsaronis studied the optimal profiles for all important control
variables in continuous cyclic distillation systems [24]. These variables included the flow rates of
feed, products, reflux and vapor. They studied two examples. In the first, an ideal ternary mixture
was separated into two fractions. In the second, the same mixture was separated into three fractions
using a column with a side stream. The optimal control problem was formulated and solved using
the software Optimal Control Code Generator for Maple (OCOMA). In the first example, the
energy demand of a column with 16 trays for separating a three-component mixture operating at
22.7% higher than minimum energy demand, was reduced by 4.2%. The study showed that the
possible energy savings were higher for columns operating at considerable higher energy demands.
25
The second example was the first study on a column with a side stream. The study showed that the
potential energy-savings can be reduced by approximately 48%. Lastly, the authors concluded that
the inclusion of oscillations in the optimal control profiles can be extended for the batch distillation
and could lead to large energy savings [24].
In 2012, Flodman and Timm studied batch distillation employing cyclic rectification and
stripping operations. The authors experimented with three modes of operation: (1) a conventional
batch still with a fractionation column configured for rectification, (2) an inverted batch still with
the column configured for stripping and (3) an operating policy where the column is sequentially
used for batch rectification and stripping. The aim for the third mode was to demonstrate a mode
that potentially could reduce operating costs by maintaining high product rates with fluids having
modest differences in volatilities. The policy was implemented on an educational batch
rectifying/stripping still equipped with a total condenser, partial reboiler, and six, 3-inch diameter
sieve trays. The control diagram for batch rectification is illustrated below in Figure 2.11.
26
Figure 2.11 Batch Rectification Control Diagram [25]
The control diagram for batch stripping is illustrated below in Figure 1.12.
Figure 2.12 Batch Stripping Control Diagram [25]
27
Conventional batch rectification and inverted batch stripping was used to promote high
product flow rates for a binary fractionation. While rectifying, the light component was removed
as distillate, concentrating the heavy component in the reboiler. As the distillate decreased with
time, the still was then switched to stripping mode. The heavy component was removed as bottoms
product, concentrating the light component in the distillate drum. For startup and for liquid
transfer, tubing allowed the bottoms to be pumped to the distillate drum, or the distillate to be
transferred to the reboiler. The mode was not optimized; however, they demonstrated energy and
time savings compared to conventional batch or inverted batch distillation alone. The advantages
of this cyclic batch operation were the simultaneous concentration for both the light and heavy
components, as well as the capability of fractionating nearly the entire contents of the initial
charge. Lastly, the authors suggested that a minimal capital investment would be required to
convert a conventional batch still to be capable of this type of cyclic operation [25].
Up to this point, most of the modeling done for cyclic distillation involved an assumption
of a linear equilibrium relationship. Lita, Bilde, and Kiss filled this gap in 2012 by modeling the
design and control of cyclic distillation systems for the general case of nonlinear equilibrium [26].
The authors developed a complete model and an innovative graphical method similar to the
McCabe-Thiele diagram, as well as demonstrated the controllability of cyclic distillation columns.
Through their study, they concluded that the energy requirements for cyclic distillation are greatly
reduced especially for high purity products. Moreover, the number of trays required is reduced by
nearly 50% when the same purity is obtained with the same vapor flow rate. The authors also
discovered that a minimum vapor-flow rate exists corresponding to an infinite number of trays, as
well as the opposite. A minimum number of trays exists with a corresponding infinite vapor flow
28
rate. Lastly, the authors suggested that columns could be easily controlled by adjusting the reflux
and the vapor flow rate to sustain required product purities [26].
Finally, in 2012 Maleta and Maleta patented the first official tray to be used for cyclic
distillation [27]. They called it a “mass exchange contact device.” The patented tray or contact
device works as follows. During the vapor period, the gas phase lifts a movable valve in such a
way that the upper plate closes the bottom, opening of the above tray. The vapor getting under the
upper plate is allowed through small orifices and enters a barbotage unit while passing through the
liquid layer therein. At the end of the vapor period, the valve moves down due to the weight of the
liquid on the above tray. The liquid from the barbotage zone on tray 1 passes through the windows
and into the liquid (which is limited by a casing) on the lower tray. The overall design of the trays
is a combination of bubble trays and sluice chambers under each tray. The liquid can move from
one tray to another without mixing of liquids from adjacent trays. A figure of the trays and a
complete column set up can be viewed below in Figure 2.13.
29
Figure 2.13 Mass Exchange Contact Device and Column [28]
The most recent study on cyclic distillation was published in 2015 [28]. Maleta,
Shevchenko, Bedryk, and Kiss reported on the performance of a pilot-scale distillation column for
ethanol-water separation operated in cyclic mode. The column had a diameter of 310 mm, height
5500 mm, and had 10 Maleta trays spaced 500 mm apart. The study was set up to feed two separate
columns at the same time using a splitter to allow for optimal performance comparison of both
operation modes. The study resulted in higher throughput and equipment productivity in the cyclic
operated column. With a beer mash feed of 4-12% ethanol, cyclic distillation had a steam usage of
1.4 times less while using 2.6 times fewer trays as compared with classical distillation. The
maximum efficiency of the process corresponded to the minimum steam consumption due to a
finite number of trays within a given column. The pilot scale distillation column was 100-160%
per the perfect mixing theoretical stage model (classis distillation), or 40-90% per the perfect
displacement model. The authors concluded that potential applications include biofuel production,
organic synthesis, specialty chemicals, gas processing, petrochemicals, and pharmaceuticals.
30
CHAPTER 3: Modeling Overview
3.1 Batch Distillation
3.1.1 Alembic Distillation – The Rayleigh Equation
The simplest type of distillation is binary distillation using no trays and consequently no
rectification, also known as Rayleigh distillation. In the spirits industry, Rayleigh distillation is
also known as alembic distillation. The pot is charged with an initial feed which is connected
directly to a condenser. As the feed is boiled, the resulting initial vapor will be rich in the most
volatile component. As the distillation continues, the concentration of the most volatile component
will decrease in the pot. Figure 3.1 is a schematic of simple batch distillation or Rayleigh
distillation.
Figure 3.1 Simple Batch Distillation Schematic
Mass balances can be written around the entire system:
𝐹 = 𝑊𝑓𝑖𝑛𝑎𝑙 + 𝐷𝑡𝑜𝑡𝑎𝑙 (3.1)
𝐹𝑥𝐹 = 𝑥𝑊,𝑓𝑖𝑛𝑎𝑙𝑊𝑓𝑖𝑛𝑎𝑙 + 𝐷𝑡𝑜𝑡𝑎𝑙𝑥𝐷,𝑎𝑣𝑔 (3.2)
31
The feed variables 𝐹 , 𝑥𝐹 are known, as well as the desired final distillate or pot concentration
either 𝑥𝑊,𝑓𝑖𝑛𝑎𝑙 or 𝑥𝐷,𝑎𝑣𝑔. Given there are 3 unknowns and 2 equations, a third equation was derived
known as the Rayleigh equation. An assumption is applied that the holdup in the accumulator and
column are negligible. Using this assumption, a differential mass balance can be applied to the
system. The differential amount of a component -dW with concentration xD is removed from the
system, resulting in the following differential mass balance:
−𝑥𝑑𝑑𝑊 = −𝑑(𝑊𝑥𝑊) = −𝑊𝑑𝑋𝑤 − 𝑥𝑊𝑑𝑊 (3.3)
Rearranging and integration gives:
𝑙𝑛 [𝑊𝑓𝑖𝑛𝑎𝑙
𝐹] = − ∫
𝑑𝑥𝑊
𝑥𝐷−𝑥𝑊
𝑥𝐹
𝑥𝑊,𝑓𝑖𝑛𝑎𝑙(3.4)
In simple batch distillations, the vapor and liquid are in equilibrium. When a total condenser is
used the substitution of 𝑦 = 𝑥𝐷 is used, resulting in:
𝑙𝑛 [𝑊𝑓𝑖𝑛𝑎𝑙
𝐹] = − ∫
𝑑𝑥
𝑦−𝑥
𝑥𝐹
𝑥𝑊,𝑓𝑖𝑛𝑎𝑙= ∫
𝑑𝑥
𝑓(𝑥)−𝑥
𝑥𝐹
𝑥𝑊,𝑓𝑖𝑛𝑎𝑙(3.5)
Here x and y are in equilibrium which can be expressed as 𝑦 = 𝑓(𝑥, 𝑝). This is the common form
of the Rayleigh equation which shows the relationship between total moles remaining in the still
and the mole fraction of the more volatile component in the still at any given time.
3.1.2 Multistage Batch Distillation
Multistage distillation refers to distillation involving one or more trays. Within these
systems the mole fraction in the distillate, 𝑥𝐷 and the mole fraction in the pot, 𝑥𝑊 are no longer in
equilibrium. Therefore, a relationship between 𝑥𝐷 and 𝑥𝑊 must be found using stage by stage
calculations. Below is a schematic of multistage distillation.
32
Figure 3.2 Multistage Batch Distillation
An assumption is made that the liquid hold up on each tray, condenser and pot is negligible.
Therefore, mass, material and energy balances can be written around stage j and at the top of the
column at any time t as follows:
𝑉𝑗+1 = 𝐿𝑗+1 + 𝐷 (3.6)
𝑉𝑗+1𝑦𝑗+1 = 𝐿𝑗𝑥𝑗 + 𝐷𝑥𝐷 (3.7)
𝑄𝐶 + 𝑉𝑗+1𝐻𝑗+1 = 𝐿𝑗ℎ𝑗 + 𝐷ℎ𝐷 (3.8)
33
When constant molal overflow for vapor, liquid and distillate is assumed, the energy balance is no
longer needed, and the component balance also known as the operating line becomes:
𝑦𝑗+1 =𝐿
𝑉𝑥𝑗 + (1 −
𝐿
𝑉) 𝑥𝐷 (3.9)
The previous equation is a straight line on an equilibrium x-y diagram, with the slope L/V and the
intercept with the y=x line at 𝑥𝐷. During batch distillation, either 𝑥𝐷 or L/V will need to vary to
satisfy the equation, and therefore the operating line. The McCabe-Thiele method for multistage
distillation with variable reflux can be seen in Figure 3.3 below.
Figure 3.3 McCabe-Thiele Diagram for Multistage Batch Distillation of Ethanol/Water
with Varibale Reflux
34
3.1.3 Stage-By-Stage Methods for Batch Rectification
To design or simulate a multicomponent batch distillation system, stage by stage
temperature, flow rates and composition profiles as a function of time are required. The
calculations are in depth but can be solved with either of two types of computer-based methods
using differential-algebraic equations which can be solved in two ways. The model presented by
Boston, is based on the multicomponent batch-rectification operation [42]. The system consists of
a partial reboiler (still-pot), a column with N equilibrium stages and a total condenser with a reflux
drum. To start the distillation, the feed is charged to the pot and heat is supplied. As the mixture
in the pot beings to vaporize it travels upwards through the column. When the vapor leaves stage
1 at the top of the column it is condensed and passes to the reflux drum. At first, a total-reflux
condition is established for a steady-state, fixed-overhead vapor flow rate. Starting at t=0, distillate
is removed from the reflux drum and accumulated in the receiving tank at a constant molar flow
rate, and a reflux ratio is established. The heat-transfer rate to the reboiler is adjusted to maintain
the overhead-vapor molar flow rate. The model is broken down into three separate sections and
model equations are derived for component material balances, a total material balance and an
energy balance. These can be seen respectively below.
For section 1:
𝑑(𝑀0𝑥𝑖,0)
𝑑𝑡= 𝑉1𝑦𝑖,1 − 𝐿0𝑥𝑖,0 − 𝐷𝑥𝑖0 (3.10)
𝑑𝑀0
𝑑𝑡= 𝑉1 − 𝐿0 − 𝐷 (3.11)
𝑄0 + 𝑑(𝑀0ℎ𝐿0
)
𝑑𝑡= 𝑉1ℎ𝑉1 − (𝐿0 + 𝐷)ℎ𝐿0
(3.12)
35
The derivative terms are accumulations due to holdup, which is assumed to be perfectly mixed. To
relate the vapor and liquid mole fractions at Stage 1, we have the phase equilibrium equation:
𝑦𝑖,1 = 𝐾𝑖,1𝑥𝑖,1 (3.13)
By combining the component material balance with the previous equation, a revised component
material balance is derived in terms of liquid-phase compositions. By combining the total material
and energy balances, a revised energy balance equation is derived that does not include 𝑑𝑀0/𝑑𝑡.
Equations for sections II and III are derived similarly. Below is the resulting model for 𝑡 = 0+,
where 𝑖 refers to the component, 𝑗 refers to the stage, and 𝑀 is the molar liquid holdup.
1. Component mole balances for the overhead-condensing system, column stages, and
reboiler, respectively:
𝑑𝑥𝑖,0
𝑑𝑡= − [
𝐿0 + 𝐷 + 𝑑𝑀0
𝑑𝑡𝑀0
] 𝑥𝑖,0 + [𝑉1𝐾𝑖,1
𝑀0] 𝑥𝑖,1 (3.14)
𝑖 = 1 𝑡𝑜 𝐶
𝑑𝑥𝑖,𝑗
𝑑𝑡= [
𝐿𝑗−1
𝑀𝑗] 𝑥𝑖,𝑗−1 − [
𝐿𝑗 + 𝐾𝑖,𝑗𝑉𝑗 +𝑑𝑀𝑗
𝑑𝑡𝑀𝑗
] 𝑥𝑖,𝑗 + [𝐾𝑖,𝑗+1𝑉𝑗+1
𝑀𝑗] 𝑥𝑖,𝑗+1 (3.15)
𝑖 = 1 𝑡𝑜 𝐶, 𝑗 = 1 𝑡𝑜 𝑁
𝑑𝑥𝑖,𝑁+1
𝑑𝑡= (
𝐿𝑁
𝑀𝑁+1) 𝑥𝑖,𝑁 − [
𝑉𝑁+1𝐾𝑖,𝑁+1 +𝑑𝑀𝑁+1
𝑑𝑡𝑀𝑁+1
] 𝑥𝑖,𝑁+1 (3.16)
𝑖 = 1 𝑡𝑜 𝐶
Where 𝐿0 = 𝑅𝐷.
2. Total mole balances for overhead-condensing system and column stages, respectively:
36
𝑉1 = 𝐷(𝑅 + 1) + 𝑑𝑀0
𝑑𝑡(3.17)
𝐿𝑗 = 𝑉𝑗+1 + 𝐿𝑗−1 − 𝑉𝑗 −𝑑𝑀𝑗
𝑑𝑡(3.18)
𝑗 = 1 𝑡𝑜 𝑁
3. Enthalpy balances around overhead-condensing system, adiabatic column stages, and
reboiler, respectively:
𝑄0 = 𝑉1(ℎ𝑉1− ℎ𝐿0
) − 𝑀0
𝑑ℎ𝐿0
𝑑𝑡(3.19)
𝑉𝑗+1 = 1
(ℎ𝑉𝑗+1− ℎ𝐿𝑗
) × [𝑉𝑗 (ℎ𝑉𝑗
− ℎ𝐿𝑗) − 𝐿𝑗−1 (ℎ𝐿𝑗−1
− ℎ𝐿𝑗) + 𝑀𝑗
𝑑ℎ𝐿𝑗
𝑑𝑡] (3.20)
𝑗 = 1 𝑡𝑜 𝑁
𝑄𝑁+1 = 𝑉𝑁+1(ℎ𝑉𝑁+1− ℎ𝐿𝑁+1
) − 𝐿𝑁(ℎ𝐿𝑁− ℎ𝐿𝑁+1
) + 𝑀𝑁+1 (𝑑ℎ𝑉𝑁+1
𝑑𝑡) (3.21)
4. Phase equilibrium on the stages and in the reboiler:
𝑦𝑖,𝑗 = 𝐾𝑖,𝑗𝑥𝑖,𝑗 (3.22)
𝑖 = 1 𝑡𝑜 𝐶, 𝑗 = 1 𝑡𝑜 𝑁 + 1
5. Mole-fractions sums at column stages and in the reboiler:
∑ 𝑦𝑖,𝑗 = ∑ 𝐾𝑖,𝑗𝑥𝑖,𝑗 = 1.0
𝐶
𝑖=1
𝐶
𝑖=1
(3.23)
𝑗 = 0 𝑡𝑜 𝑁 + 1
6. Molar holdups in the condenser system and on the column stages, based on constant-
volume holdups, 𝐺𝑗.
𝑀0 = 𝐺0𝜌0 (3.24)
37
𝑀𝑗 = 𝐺𝑗𝜌𝑗 , 𝑗 = 1 𝑡𝑜 𝑁 (3.25)
Where ρ is liquid molar density.
7. Variation of molar holdup in the reboiler, where 𝑀𝑁+10 is the initial charge to the reboiler.
𝑀𝑁+1 = 𝑀𝑁+10 − ∑ 𝑀𝑗 − ∫ 𝐷 𝑑𝑡
𝑡
0
𝑁
𝑗=0
(3.26)
The previous described equations make up an initial-value problem for a system of ordinary
differential and algebraic equations. The total number of equations is (2𝐶𝑁 + 3𝐶 + 4𝑁 +
7). If variables 𝑁, 𝐷, 𝑅 =𝐿0
𝐷, 𝑀𝑁+1
0 , and all 𝐺𝑗 are specified, and if correlations are
available for computing liquid densities, vapor and liquid enthalpies, and K-values, the
number of unknown variables, distributed as follows, is equal to the number of equations.
Table 3.1 Variables and Total Number of Equations for Multicomponent Batch Distillation
𝑥𝑖,𝑗 𝐶𝑁 + 2𝐶
𝑦𝑖,𝑗 𝐶𝑁 + 𝐶
𝐿𝑗 𝑁
𝑉𝑗 𝑁 + 1
𝑇𝑗 𝑁 + 2
𝑀𝑗 𝑁 + 2
𝑄0 1
𝑄𝑁+1 1
2𝐶𝑁 + 3𝐶 + 4𝑁 + 7
Initial values at 𝑡 = 0 for all the above variables are obtained from steady-state, total reflux
calculations which depend on the values of 𝑁, 𝑀𝑁+10 , 𝑥𝑁+1
0 , 𝐺𝑗 , and 𝑉1. The previous equations in
section 1, 2 and 3 include first derivatives of 𝑥𝑖,𝑗, 𝑀𝑗 , and ℎ𝐿𝑗 . With the exception of 𝑀𝑁+1,
derivatives of the other two variables can be approximated by incremental changes over the
38
previous time step. If the reflux ratio is high, 𝑀𝑁+1 may also be approximated. This reduces the
ODE’s to only 𝐶(𝑁 + 2) equations for the component material balances to be integrated in terms
of the 𝑥𝑖,𝑗 dependent variables.
3.1.4 Governing Equations and Thermodynamic Model for System in Study
Applying the previous modeling method provides differential and algebraic equations which will
be used to model the system in study. Below is schematic of the column and flows in study.
Figure 3.4 Schematic of Cyclic Distillation on Carl Still
The partial condensers are assumed to act like stages and will be modeled as such. The material
balance equations for the stages (N=1-5) in the column is:
𝑑𝑥𝑁
𝑑𝑡=
𝐿(𝑥𝑁−1 − 𝑥𝑁) + 𝑉(𝑦𝑁+1 − 𝑦𝑁)
𝑀𝐻
(3.27)
The material balance for the condenser is:
39
𝑑𝑥0
𝑑𝑡=
𝑉𝑦1
𝑀𝐻0−
(𝐿 + 𝐷)𝑥0
𝑀𝐻0
(3.28)
The material balance for the reboiler is:
𝑑𝑥𝐵
𝑑𝑡= 𝐿(𝑥5 − 𝑥𝐵) − 𝑉(𝑦𝐵 − 𝑥𝐵) (3.29)
A solution of water and ethanol is considered as a non-ideal mixture because it presents an
azeotrope at 89% mole fraction of ethanol. Therefore, modified Raoult’s law was used to determine
the vapor mole fraction given a temperature and liquid mole fraction.
𝑦𝑖𝑃 = 𝑥𝑖𝛾𝑃𝑖𝑠𝑎𝑡 (3.30)
Where 𝛾 is an activity coefficient that can be calculated by using a specific thermodynamic model.
In this study, the NRTL method was the thermodynamic model chosen based on a study which
reported that the NRTL method fit experimental data accurately to an RMS value of 0.403% [41].
It was important to pick a model that accurately modeled the azeotrope between water and ethanol.
The NRTL activity model is calculated using parameters 𝐺𝑖𝑗 and 𝜏𝑖𝑗:
𝑙𝑛(𝛶1) = 𝑥22 ∗ (𝜏21 ∗ (
𝐺21
𝑥1 + 𝑥2 ∗ 𝐺21)
2
+ 𝜏12 ∗𝐺12
(𝑥2 + 𝑥1 ∗ 𝐺12)2) (3.31)
𝑙𝑛(𝛶1) = 𝑥12 ∗ (𝜏12 ∗ (
𝐺12
𝑥2 + 𝑥1 ∗ 𝐺12)
2
+ 𝜏21 ∗𝐺21
(𝑥1 + 𝑥2 ∗ 𝐺21)2) (3.32)
Where 𝐺𝑖𝑗 = exp(−0.3 ∗ 𝜏𝑖𝑗) , 𝜏𝑖𝑗 =𝐴𝑖𝑗
𝑅𝑇
The parameters 𝐴12 and 𝐴21 were found within literature to be -633 and 5823.1, respectively [41].
To determine the vapor pressure the Antoine equation was used:
40
𝑙𝑜𝑔10 𝑃𝑠𝑎𝑡 = 𝐴 − 𝐵
𝑇+𝐶(3.33)
Where pressure is in mmHg and temperature is in is in °C. Finally, the total pressure was calculated
as the sum of the partial pressures using the follow equation:
𝑃 = 𝑥1Υ1𝑃1𝑠𝑎𝑡 + 𝑥2Υ2𝑃2
𝑠𝑎𝑡 (3.34)
3.2 Cyclic Continuous Distillation Modeling Approach
As previously stated, cyclic column operation consists of two parts, a vapor flow period,
and a liquid flow period. Since the operation is two separate parts, the mathematical analysis must
be done in two parts as well. Sections 3.2.1 – 3.2.5 describe the method first published by Lita,
Bilde and Kiss for the modeling of continuous cyclic distillation systems and comparison against
conventional distillation.
3.2.1 Assumptions
The current accepted modeling approach is derived using the following assumptions:
• Binary (mixture) distillation
• Ideal stages (vapor-liquid equilibrium is reached)
• Equal heat of vaporization (constant molar holdup and vapor flow rate)
• Perfect mixing on each stage
• Negligible vapor holdup
• Saturated liquid feed
3.2.2 Operational Constraints
From the condenser and reboiler mass balance, for one operating cycle:
41
𝑉 ∗ 𝑡𝑣𝑎𝑝 = 𝐷 + 𝐿 (3.35)
𝐿 + 𝐹 = 𝑉 ∗ 𝑡𝑣𝑎𝑝 + 𝐵 (3.36)
the condition follows that:
𝐿 < 𝑉 ∗ 𝑡𝑣𝑎𝑝 < 𝐿 + 𝐹 (3.37)
3.2.3 Model of Vapor Flow Period
The model for the vapor flow period involves using equations to describe the evolution in time of
stage holdup and composition.
𝐶𝑜𝑛𝑑𝑒𝑛𝑠𝑒𝑟: 𝑑𝑀1 𝑉∗𝑡𝑣𝑎𝑝=𝐷+𝐿
𝑑𝑡= 𝑉; 𝑀1
𝑑𝑥1
𝑑𝑡= 𝑉(𝑦2 − 𝑥1) (3.38)
𝑇𝑟𝑎𝑦𝑠: 𝑑𝑀𝑘
𝑑𝑡= 0; 𝑀𝑘
𝑑𝑥𝑘
𝑑𝑡= 𝑉(𝑦𝑘+1 − 𝑦𝑘) (3.39)
𝑅𝑒𝑏𝑜𝑖𝑙𝑒𝑟: 𝑑𝑀𝑁𝑇
𝑑𝑡= −𝑉; 𝑀𝑁𝑇
𝑑𝑥𝑁𝑇
𝑑𝑡= 𝑉(𝑥𝑁𝑇 − 𝑦𝑁𝑇) (3.40)
𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝑐𝑜𝑛𝑑𝑖𝑡𝑖𝑜𝑛𝑠: 𝑎𝑡 𝑡 = 0, (𝑀, 𝑥) = (𝑀(𝐿), 𝑥(𝐿)) (3.41)
Integration of these equations from 𝑡 = 0 to 𝑡 = 𝑡𝑣𝑎𝑝 gives the state of the system at the end of the
vapor flow period, (𝑀(𝑉), 𝑥(𝑉)) [26].
3.2.4 Model of Liquid Flow Period
The model for the liquid flow period uses the following equations:
𝐶𝑜𝑛𝑑𝑒𝑛𝑠𝑒𝑟: 𝑀1(𝐿)
= 𝑀1(𝑉)
− 𝐷 − 𝐿; 𝑥1(𝐿)
= 𝑥1(𝑉) (3.42)
𝑇𝑟𝑎𝑦𝑠, 𝑟𝑒𝑐𝑡𝑖𝑓𝑦𝑖𝑛𝑔 𝑠𝑒𝑐𝑡𝑖𝑜𝑛: 𝑀𝑘(𝐿)
= 𝐿; 𝑥𝑘(𝐿)
= 𝑥𝑘−1(𝑉) (3.43)
42
𝐹𝑒𝑒𝑑 𝑡𝑟𝑎𝑦: 𝑀𝑁𝐹+1(𝐿)
= 𝐿 + 𝐹; 𝑥𝑁𝐹+1(𝐿)
=𝐿𝑥𝑁𝐹
(𝑉)+ 𝐹𝑥𝐹
(𝐿 + 𝐹)(3.44)
𝑅𝑒𝑏𝑜𝑖𝑙𝑒𝑟: 𝑀𝑁𝑇(𝐿)
= 𝑀𝑁𝑇(𝑉)
− 𝐵 + 𝑀𝑁𝑇−1(𝑉)
; 𝑥𝑁𝑇(𝐿)
=(𝑀𝑁𝑇
(𝑉)− 𝐵)𝑥𝑁𝑇
(𝑉)+ 𝑀𝑁𝑇−1
(𝑉)𝑥𝑁𝑇−1
(𝑉)
𝑀𝑁𝑇(𝐿)
(3.45)
3.2.5 Solution Method
For the solution, the previous equations are written in condensed form, where Φ(𝑉) and Φ(𝐿) are
mappings which relate the state at the beginning and at the end of the liquid and vapor flow periods,
respectively:
(𝑀(𝑉), 𝑥(𝑉)) = Φ(𝑉)(𝑀, 𝑥) (3.46)
(𝑀(𝐿), 𝑥(𝐿)) = Φ(𝐿)(𝑀, 𝑥) (3.47)
Which implies the following condition:
(𝑀(𝐿), 𝑥(𝐿)) = Φ(𝐿) ∘ Φ(𝑉)(𝑀(𝐿), 𝑥(𝐿)) (3.48)
A solution to the above equation can be found by considering an initial state and applying
relationships for (𝑀(𝑉), 𝑥(𝑉)) and (𝑀(𝐿), 𝑥(𝐿)) until the difference between two iterations becomes
small. Alternatively, the convergence of the iterations can also be done by applying numerical
methods and can be solved in MATLAB © [26]. Below is a graphically representation of the
energy requirements in cyclic distillation compared to classic distillation.
43
Figure 3.5 Comparison of Energy Requirements in Cyclic Distillation versus Conventional
Distillation [26]
The vapor/feed ratio is comparable to the energy requirements of a distillation column. The
higher the ratio, the more energy required. 1 − 𝑥𝐷 represents the product purity. The energy
requirements are greatly reduced and is reduced even greater when the product purity is higher.
3.3 Design of Cyclic Distillation
3.3.1 Design Methodology
First given the feed (𝐹, 𝑥𝐹) and the required performance (𝑥𝐷 , 𝑥𝐵) the mass balance over one
operating cycle is used to find the product flow rates (𝐷, 𝐵).
𝐹 = 𝐷 + 𝐵 (3.49)
𝐹 ∗ 𝑥𝐹 = 𝐷 ∗ 𝑥𝐷 + 𝐵 ∗ 𝑥𝐵 (3.50)
Next the vapor flow rate 𝑉and the duration of the vapor-flow period, 𝑡𝑣𝑎𝑝, are specified. The liquid
transferred from the condenser to the top tray can be calculated.
44
𝐿 = 𝑉 ∗ 𝑡𝑣𝑎𝑝 − 𝐷 (3.51)
For tray holdups in rectifying section 𝑀𝑘 = 𝐿 and for the stripping section we have 𝑀𝑘 = 𝐿 + 𝐹.
It is important to check hydrodynamics of the column including, column diameter, vapor velocity
and pressure drop.
To determine the state of the reboiler at the end of the vapor-flow period the following needs to be
specified:
- Holdup, 𝑀𝑁𝑇(𝑡𝑣𝑎𝑝). Although the results of this design procedure are independent of this
value it is still necessary to specify.
- Composition, 𝑥𝑁𝑇(𝑡𝑣𝑎𝑝) = 𝑥𝐵. At the end of the vapor-flow period, the reboiler is richer
in the heavy component and therefore is the time to remove the bottom product if possible.
To find the holdup and reboiler composition at the beginning of the vapor-flow period,
𝑀𝑁𝑇(0), 𝑥𝑁𝑇(0), equations (3.32) are integrated from 𝑡 = 𝑡𝑣𝑎𝑝 to 𝑡 = 0. Next determine the state
of the last tray (stage NT-1) at the end of the vapor flow period, using the mass balance for the
liquid flow period:
𝑀𝑁𝑇−1(𝑡𝑣𝑎𝑝) = 𝐿 + 𝐹 (3.52)
𝑥𝑁𝑇−1(𝑡𝑣𝑎𝑝) =𝑀𝑁𝑇(0) ∗ 𝑥𝑁𝑇(0) − (𝑀𝑁𝑇(𝑡𝑣𝑎𝑝) − 𝐵) ∗ 𝑥𝑁𝑇(𝑡𝑣𝑎𝑝)
𝐿 + 𝐹(3.53)
Next find the state of the reboiler and last tray at the beginning of the vapor flow period by
integrating equations 3.30 and 3.31 from 𝑡 = 𝑡𝑣𝑎𝑝 𝑡𝑜 𝑡 = 0. Next add one more tray, whose state
at the end of the vapor-flow period is
45
𝑀𝑁𝑇−2(𝑡𝑣𝑎𝑝) = 𝐿 + 𝐹 (3.54)
𝑥𝑁𝑇−2(𝑡𝑣𝑎𝑝) = 𝑥𝑁𝑇−1(0) (3.55)
And integrate the resulting set of equations from 𝑡 = 𝑡𝑣𝑎𝑝 𝑡𝑜 𝑡 = 0. Repeat until the feed
composition is reached for the tray NF+1. Then find the state of the feed tray at the end of the
vapor flow period.
𝑀𝑁𝐹(𝑡𝑣𝑎𝑝) = 𝐿 (3.56)
𝑥𝑁𝐹(𝑡𝑣𝑎𝑝) =𝑀𝑁𝐹+1∗𝑥𝑁𝐹+1(0)−𝐹∗𝑥𝐹
𝑀𝑁𝐹(3.57)
And integrate from 𝑡 = 𝑡𝑣𝑎𝑝 𝑡𝑜 𝑡 = 0. Similarly, repeat addition of one tray, finding its state at
the end of the vapor-flow period and integration of the resulting equations until the distillate
composition is reached. The previously discussed process is illustrated in Figure 3.5 below.
Figure 3.6 Cyclic Distillation Design Approach
3.4 Batch Cyclic Distillation Dynamic Analysis
The basis of the dynamic analysis of batch cyclic distillation comes from McWhirter’s
dissertation from the 1960’s. The equations developed were written for a continuously fed batch
system. Here the equations are re-written to apply to batch distillation where necessary. Below is
a schematic of the distillation column used for cyclic operation in the lab. As seen in Figure 3.6
below, in terms of dynamic modeling partial condensers act similarly to stages therefore the partial
46
condensers are modeled as such. The following sections will provide the equations written to
model the system in study.
3.4.2 Vapor Flow Period
During the vapor flow period, only vapor flows up through the partial condensers and the
column and ideally no liquid flows down the column. The column liquid hold up is trapped and
held on the stages in the column during this vapor flow period. A detailed analysis will be made
for stage N. The material balance for the most volatile component on stage N is written as:
𝑑(𝑈𝑁𝑌𝑁)
𝑑𝑡+
𝑑(𝑊𝑁𝑋𝑁)
𝑑𝑡= 𝑉𝑁+1𝑌𝑁+1 − 𝑉𝑁𝑌𝑁 (3.58)
Where,
𝑈𝑁 = 𝑣𝑎𝑝𝑜𝑟 ℎ𝑜𝑙𝑑 𝑢𝑝 𝑜𝑛 𝑠𝑡𝑎𝑔𝑒 𝑁, 𝑚𝑜𝑙𝑒𝑠
𝑊𝑁 = 𝑙𝑖𝑞𝑢𝑖𝑑 ℎ𝑜𝑙𝑑𝑢𝑝 𝑜𝑛 𝑠𝑡𝑎𝑔𝑒 𝑁, 𝑚𝑜𝑙𝑒𝑠
𝑌𝑁 = 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑣𝑎𝑝𝑜𝑟 𝑚𝑜𝑙𝑒 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑜𝑓 𝑚𝑜𝑟𝑒 𝑣𝑜𝑙𝑎𝑡𝑖𝑙𝑒 𝑐𝑜𝑚𝑝𝑜𝑛𝑒𝑛𝑡 𝑙𝑒𝑎𝑣𝑖𝑛𝑔 𝑠𝑡𝑎𝑔𝑒 𝑁
𝑋𝑁 = 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑙𝑖𝑞𝑢𝑖𝑑 𝑚𝑜𝑙𝑒 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑜𝑓 𝑚𝑜𝑟𝑒 𝑣𝑜𝑙𝑎𝑡𝑖𝑙𝑒 𝑐𝑜𝑚𝑝𝑜𝑛𝑒𝑛𝑡 𝑜𝑛 𝑠𝑡𝑎𝑔𝑒 𝑁
𝑉𝑁 = 𝑣𝑎𝑝𝑜𝑟 𝑓𝑙𝑜𝑤 𝑟𝑎𝑡𝑒 𝑓𝑟𝑜𝑚 𝑠𝑡𝑎𝑔𝑒 𝑁, 𝑚𝑜𝑙𝑒𝑠/𝑚𝑖𝑛
𝑡 = 𝑡𝑖𝑚𝑒 𝑖𝑛 𝑚𝑖𝑛𝑠
The energy balance relationship is then:
𝑑(𝑈𝑁𝐻𝑁)
𝑑𝑡+
𝑑(𝑊𝑁ℎ𝑁)
𝑑𝑡= 𝑉𝑁+1𝐻𝑁+1 − 𝑉𝑁𝐻𝑁 − 𝑄𝑁 (3.59)
Where,
47
𝐻𝑁 = 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑒𝑛𝑡ℎ𝑎𝑙𝑝𝑦 𝑜𝑓 𝑣𝑎𝑝𝑜𝑟 𝑙𝑒𝑎𝑣𝑖𝑛𝑔 𝑠𝑡𝑎𝑔𝑒 𝑁, 𝑘𝐽 𝑝𝑒𝑟 𝑚𝑜𝑙𝑒
ℎ𝑁 = 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑒𝑛𝑡ℎ𝑎𝑙𝑝𝑦 𝑜𝑓 𝑙𝑖𝑞𝑢𝑖𝑑 𝑜𝑛 𝑠𝑡𝑎𝑔𝑒 𝑁, 𝑘𝐽 𝑝𝑒𝑟 𝑚𝑜𝑙𝑒
𝑄𝑁 = ℎ𝑒𝑎𝑡 𝑙𝑜𝑠𝑠 𝑓𝑟𝑜𝑚 𝑠𝑡𝑎𝑔𝑒 𝑁, 𝑘𝐽 𝑝𝑒𝑟 ℎ𝑜𝑢𝑟
The average enthalpies are a function of the liquid mole fraction and temperature on stage N. For
simplicity, the difference between enthalpies (𝐻𝑁+1 − ℎ𝑁) is assumed independent of
composition. This allows for the assumption of constant value of ℎ𝑁 on each stage. Also, the vapor
hold up is assumed negligible in comparison to liquid hold up. With these assumptions the energy
balance becomes:
𝑑𝑊𝑁
𝑑𝑡= 0 (3.60)
The liquid holdup per stage during the vapor flow period is constant and equal for all stages. Using
this and the previous assumptions, the material balance becomes:
𝑑𝑋𝑁
𝑑𝑡=
𝑉(𝑌𝑁+1 − 𝑌𝑁)
𝑊(3.61)
The previous equation completely describes the material balance and must be written for each
stage in the column of study and all must be satisfied.
The reflux to the column is assumed to be cycled in phase with the liquid flow period.
Also, the condenser is assumed to have a holdup equal to the amount of vapor introduced into it
during the vapor flow period. Consequently, the entire contents of the condenser will be returned
as reflux or withdrawn as distillate product during each cycle.
The overall material balance expression for the reboiler is:
48
𝑑𝑊𝐵
𝑑𝑡+
𝑑𝑈𝐵
𝑑𝑡= −𝑉 (3.62)
Where,
𝑊𝐵 = 𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 𝑙𝑖𝑞𝑢𝑖𝑑 ℎ𝑜𝑙𝑑𝑢𝑝, 𝑝𝑜𝑢𝑛𝑑 𝑚𝑜𝑙𝑒𝑠
𝑈𝐵 = 𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 𝑣𝑎𝑝𝑜𝑟 ℎ𝑜𝑙𝑑𝑢𝑝, 𝑝𝑜𝑢𝑛𝑑 𝑚𝑜𝑙𝑒𝑠
𝑉 = 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 𝑚𝑜𝑙𝑎𝑟 𝑣𝑎𝑝𝑜𝑟 𝑟𝑎𝑡𝑒, 𝑝𝑜𝑢𝑛𝑑 𝑚𝑜𝑙𝑒𝑠 𝑝𝑒𝑟 ℎ𝑜𝑢𝑟
The contents of the reboiler is assumed to be perfectly mixed. The material balance on the most
volatile component is:
𝑑(𝑊𝐵𝑥𝐵)
𝑑𝑡+
𝑑(𝑈𝐵𝑌𝐵)
𝑑𝑡= −𝑉𝑌𝐵 (3.63)
Similarly, the vapor hold up is considered negligible in comparison to the liquid hold up. Using
this assumption and substituting the material balance becomes:
𝑑𝑥𝐵
𝑑𝑡=
−𝑉
𝑤𝐵
(𝑌𝐵 − 𝑥𝐵) (3.64)
Adding equations for the reboiler to the previous stage equations completely models the vapor
flow period for the system in study.
3.4.3 Liquid Flow Period
During the liquid flow period ideally only liquid flows down the column to the preceding
tray and there is no vapor traveling up the column. Since the relative velocity between the phases
will be almost zero and the vapor hold up is essentially negligible, no mass transfer will be assumed
to take place during the liquid flow period. Consequently, time is not a factor in any calculations
for the liquid flow period. Plug flow of the liquid from stage to stage down the column is assumed.
49
The volume of liquid that flows from stage to stage is assumed equal for all stages and constant
for all liquid flow periods. The following equation gives a material balance for stage N:
𝑥𝑁𝐶+1 =
𝑂𝐿𝑥𝑁−1𝐶
𝑊+
(𝑊 − 𝑂𝐿)
𝑊𝑥𝑁
𝐶 (3.65)
Where,
𝑂𝐿
= 𝑞𝑢𝑎𝑛𝑡𝑖𝑡𝑦 𝑜𝑓 𝑙𝑖𝑞𝑢𝑖𝑑 𝑤ℎ𝑖𝑐ℎ 𝑓𝑙𝑜𝑤𝑠 𝑓𝑟𝑜𝑚 𝑠𝑡𝑎𝑔𝑒 𝑡𝑜 𝑠𝑡𝑎𝑔𝑒 𝑑𝑢𝑟𝑖𝑛𝑔 𝑡ℎ𝑒 𝑙𝑖𝑞𝑢𝑖𝑑 𝑓𝑙𝑜𝑤 𝑝𝑒𝑟𝑖𝑜𝑑, 𝑙𝑏 𝑚𝑜𝑙𝑒𝑠
𝑐 = 𝑐𝑦𝑐𝑙𝑒 𝑛𝑢𝑚𝑏𝑒𝑟
The material balance calculates the value of the concentration of the liquid on stage N at the start
of a new vapor flow period from the values at the end of the preceding vapor flow period. To
simplify Equation 3.60, a new quantity is introduced, ∅, which is the ratio of the quantity of liquid
which flows during the liquid flow period to the liquid holdup per stage. Using this variable,
Equation 3.60 becomes:
𝑥𝑁𝐶+1 = ∅𝑥𝑁−1
𝐶 + (1 − ∅)𝑥𝑁𝐶 (3.66)
The previous equation is valid for values of ∅ from zero to one, but when values are larger than
one the following will apply:
For 1.0 ≤ ∅ ≤ 2.0
𝑥𝑁𝐶+1 = (∅ − 1.0)𝑥𝑁−2
𝐶 + (2.0 − ∅)𝑥𝑁−1𝐶 (3.67)
For 2.0 ≤ ∅ ≤ 3.0
𝑥𝑁𝐶+1 = (∅ − 2.0)𝑥𝑁−3
𝐶 + (3.0 − ∅)𝑥𝑁−2𝐶 (3.68)
50
During the liquid flow period, no vapor leaves the reboiler, but reflux from the partial condenser
is re-introduced back into the reboiler. The average composition of the reflux from the partial
condenser depends on the value of ∅.
For 0.0 ≤ ∅ ≤ 1.0
𝑥𝐿 = 𝑥𝑁𝑂𝑆 (3.69)
Where 𝑥𝐿 = 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑐𝑜𝑚𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛 𝑜𝑓 𝑟𝑒𝑓𝑙𝑢𝑥 𝑓𝑟𝑜𝑚 𝑏𝑜𝑡𝑡𝑜𝑚 𝑜𝑓 𝑐𝑜𝑙𝑢𝑚𝑛
𝑥𝑁𝑂𝑆 = 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑐𝑜𝑚𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛 𝑜𝑓 𝑙𝑖𝑞𝑢𝑖𝑑 𝑜𝑛 𝑏𝑜𝑡𝑡𝑜𝑚 𝑠𝑡𝑎𝑔𝑒 𝑎𝑡 𝑒𝑛𝑑 𝑜𝑓 𝑣𝑎𝑝𝑜𝑟 𝑓𝑙𝑜𝑤 𝑝𝑒𝑟𝑖𝑜𝑑
For 1.0 ≤ ∅ ≤ 2.0
𝑥𝐿 = (∅ − 1.0)𝑥𝑁𝑂𝑆−1
∅+
𝑥𝑁𝑂𝑆
∅(3.70)
For 2.0 ≤ ∅ ≤ 3.0
𝑥𝐿 = (∅ − 2.0)𝑥𝑁𝑂𝑆−2
∅+
𝑥𝑁𝑂𝑆−1 + 𝑥𝑁𝑂𝑆
∅(3.71)
The material balance for the most volatile component for the reboiler during the liquid flow period
is given below:
𝑥𝐵𝐶+1 =
𝑊𝐵𝑥𝐵𝐶
𝑊𝐵𝐼+
∅(𝑊)(𝑥𝐿)
𝑊𝐵𝐼
(3.72)
Where 𝑊𝐵𝐼 = 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 ℎ𝑜𝑙𝑑𝑢𝑝 𝑎𝑡 𝑠𝑡𝑎𝑟𝑡 𝑜𝑓 𝑣𝑎𝑝𝑜𝑟 𝑓𝑙𝑜𝑤 𝑝𝑒𝑟𝑖𝑜𝑑, 𝑝𝑜𝑢𝑛𝑑 𝑚𝑜𝑙𝑒𝑠
𝑥𝐵𝐶+1 = 𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 𝑐𝑜𝑚𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛 𝑎𝑡 𝑠𝑡𝑎𝑟𝑡 𝑜𝑓 𝑛𝑒𝑥𝑡 𝑣𝑎𝑝𝑜𝑟 𝑓𝑙𝑜𝑤 𝑝𝑒𝑟𝑖𝑜𝑑
51
Equations 3.53 through 3.65 describe the operation of the column throughout one entire cycle. The
equations with correct subscripts can be found Appendix B.
3.4.4 Solution Method
Like how the continuous equations are solved, each cycle requires the end point of the
previous cycle. Therefore, to start the vapor flow period equations are used to solve with initial
conditions for the state of the system at the end of vapor flow period 1. Those values are then used
to solve for the state of the system at the end of liquid flow period 1. Those numbers are then used
to solve for the state of the system at the end of vapor flow period 2. This process continues until
the final condenser ABV drops below 10%. The MATLAB script written for the solution and
modeling of cyclic batch distillation can be found in Appendix B. The results can be found in
Appendix B and Chapter 5, Results.
52
CHAPTER 4: Methods
4.1 Materials and Equipment
Raw materials used for this research included: fermented apple cider, previously distilled
apple brandy, standard solutions, and cleaning agents. Equipment used for this research included:
165 L Carl batch still, leur lock syringes, copper tubing, centrifugal pump, GC-2010 gas
chromatograph, plastic GC vials, storage tanks, hydrometer, and miscellaneous glassware.
4.1.1 Manufacturing Equipment
To monitor the tray composition during distillation, four site glasses on the distillation
column where replaced with a new piece of glass made of borosilicate, which had a precut hole to
allow for a sampling apparatus to be installed. Borosilicate was chosen due to its high melting
point and non-reactivity with the compounds being distilled. The sampling apparatus consisted of
a luer-lock system with copper tubing going to the column and resting on the tray. Outside of the
column a syringe allowed a 1 mL sample to be extracted from the tray. A schematic of the self-
manufactured sampling equipment can be seen below.
53
Figure 4.1 Manufactured Sampling Apparatus
In addition to the fabrication of sample ports, an extra valve between the steam condensate
outlet to the drain was installed to allow for the collection of the condensate. This allowed for the
condensate to be collected into buckets, transported to a larger container and measured. The
volume was then converted to pounds of steam and reported as pounds of steam in the data.
4.2 Measurement Methods
4.2.1 GC Method
A gas chromatograph was used to analyze the samples withdrawn from the still. Below is
a list of compounds that were of importance and their respective boiling points.
54
Table 4.1 Detected Compounds [44]
Compound Boiling Point
(°C)
Retention TIME
(MIN)
Acetaldehyde 20.8 1.532
Acetone 56.2 2.194
Ethyl Acetate 77.0 2.913
Methanol 64.7 3.091
Isopropanol 82.6 3.509
Ethanol 78.0 3.615
Propanol 97.0 5.184
Isobutanol 108 6.132
Butanol 117.7 6.944
Isoamyl Alcohol 132.0 7.887
The above compounds were selected based on previous knowledge of the compounds metabolized
by common yeast strains used in fermented beverage production and are volatile enough to come
through to the final product in distilled alcoholic beverages. Although there are many other
compounds that yeast produce including aldehydes, ketones, and esters, the compounds chosen to
have the most impact on the flavor and aroma profile of sprits. Sample chromatograph results can
be seen in Figure C.1 in Appendix C.
Standards with four different levels of concentrations ranging from 0.1 – 10 g/L were made
to form a method for the GC-2010 gas chromatograph. The column used was a 30 m Stabliwax
column with 0.32 mm ID and film thickness of 0.50 um. The oven temperature started at 35 °C
and ramping up to 100 °C over a period of 8.5 minutes. The carrier gas used was helium with a
linear velocity of 45 cm/sec and a split ratio of 50. The auto injector AOC-20i was paired with the
GC to analyze 12 samples in one batch.
55
4.3 Research Design
Multiple distillations were carried out to understand how cyclic distillation compares to
conventional distillation in the 165 L batch still. Table 4.2 shows all distillations performed and
the operating conditions associated with each distillation including: the duration of the liquid and
vapor flow periods during cyclic operation, the initial volume and ABV, and whether reflux to the
partial condensers was being utilized or not.
Table 4.2 Distillation Conditions
Distillation # Operation Type
(Vapor, Liquid)
Initial
Volume
(gals)
Initial ABV Reflux to
Partial
Condensers
1 Conventional 47 35% Off
2 Conventional 48 25% Off
3 Conventional 49 40% Off
4 Cyclic
(9 min, 3 min)
47 36% Off
5 Cyclic
(6 min, 3 min)
45 38% Off
6 Cyclic
(3 min, 3 min)
39 36% Off
7 Conventional 25 8% Off
8 Cyclic
(9 min, 3 min)
25 8% Off
9 Cyclic
(6 min, 3 min)
25 7.4% Off
10 Cyclic
(4 min, 3 min)
25 6.3% Off
11 Conventional 36 19% On
12 Conventional 36 27% Off
13 Cyclic
(6 min, 3 min)
Tray by Tray
36 28% On
14 Cyclic
(6 min, 3min)
Tray by Tray
33 27% On
15 Cyclic
(6 min, 3min)
Tray by Tray
33 30% On
56
4.4 Procedures
To begin a distillation, the pot was filled with the intended initial volume. The door was
then shut tightly, and steam valves opened. The condensate valve to the steam jacket was then
opened to release any water remaining in the jacket from the previous distillation and then closed.
In the beginning of every distillation, steam was supplied to the steam jacket until the first volume
of distillate was collected. The distillate flowrate and pressure reading were monitored throughout
the distillation throughout each distillation. If the flowrate dropped below average levels, the
steam was increased using the gate valve to keep a relatively constant distillate flowrate. During
conventional operation, steam was supplied to the column for the entire duration of the distillation,
until the hydrometer read below 10% ABV.
During cyclic distillation, the first vapor flow period started when the first distillate was
collected. After the appropriate vapor flow time, the main steam valve was closed. This started
the liquid flow period. During the liquid flow period, the tray valves were opened to allow the
liquid to flow to the tray below. The bottom tray was first opened for around 10-15 seconds, until
the stream coming down from the tray nearly stopped. Next, the valve was closed, and the middle
tray was opened. Lastly, the top tray valve was opened to allow the contents to fall to the middle
tray and then closed again. The distillation continued until the hydrometer read under 10% ABV
at the beginning of a vapor flow period.
4.5 Data Collection and Analysis
Throughout each distillation temperature, pressure, distillate flow rate, and total volume
collected were taken at time increments roughly 10 minutes apart. Samples from each tray, the
bottom of the column, and of the distillate were taken roughly 10 minutes apart and taken during
57
each vapor period for cyclic operation. The samples were placed directly into GC vials and labeled.
Steam condensate was collected into buckets and stored in a large storage tank to record overall
volume after the distillation was complete. The distillate flow rate was measured using a stopwatch
and graduated cylinder, collected distillate over a 10 second period and recorded as mL/s. The
distillate was collected into volume incremented containers. After every distillation was complete
samples were taken of the final contents of the pot and of the overall distillate, as well as of cuts if
they were taken.
After the distillation was complete, each sample was run through the GC-2010 gas
chromatograph to analyze the concentration of each component in each sample. The results can be
seen in Chapter 5.
58
CHAPTER 5: Results
5.1 Results
A complete set of graphical results for every distillation and every compound studied can
be found in Appendix A. Within this section are key results which will be discussed and referred
to in the discussion section
5.1.1 All Distillation Results
Table 5.1 is a summary table of all the distillations, included in it is the initial parameters,
% recovered, and steam and time efficiencies.
Table 5.1 Distillation Efficiency Results
Figure 5.1 shows the hydrometer reading versus time throughout every distillation. D1 is not
shown because not enough data was collected.
Distillation # Type (V,L) Reflux Initial Distilled Recovered Final Pot Steam Used Efficiency Time Time Eff.
PG PG % ABV % lbs lbs steam/ PG hours PG/hr
3 Normal Operation Off 39 34 0.87 1.2% 284 8.4 4.1 8.3
4 Cyclic (9 min, 3 min) Off 34 32 0.95 0.6% 266 8.3 4.5 7.1
5 Cyclic(6 min, 3 min) Off 34 31 0.90 1.3% 242 7.9 5.9 5.2
6 Cyclic(4 min, 3 min) Off 28 27 0.96 2.0% 223 8.3 6.0 4.5
7 Normal Operation - Low Wines Off 4 4 0.90 0.50% 114 31.7 1.3 2.8
8 Cyclic -Low Wines (9min, 3 min) Off 4 4 0.88 0.70% 116 33.1 2.2 1.6
9 Cyclic - Low Wines (6min, 3min) Off 4 4 0.99 0.40% 126 34.3 3.3 1.1
10 Cyclic - Low Wines (4min, 3min) Off 3 3 0.96 0.40% 134 44.2 3.2 0.9
11 Normal Operation (Brandy) On 14 13 0.92 0.20% 183 14.2 2.2 6.0
12 Normal Operation (Brandy) Off 20 18 0.92 0.80% 175 9.7 2.6 6.9
13 Cyclic (6min, 3min) - Tray by Tray On 20 19 0.92 0.20% 246 13.1 3.6 5.2
14 Cyclic (6min, 3min) - Tray by Tray On 18 17 0.93 0.20% 225 13.5 2.7 6.1
15 Cyclic (6min, 3min) - Tray by Tray On 20 18 0.92 0.20% 224 12.3 2.8 6.5
59
Figure 5.1 Ethanol Concentration in Distillate vs Time for All Distillations
Figure 5.2 shows the hydrometer reading vs volume for distillations D7 through D15. These
distillations are the only distillations in which volume distilled data was collected, so they were
selected for comparison.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 50 100 150 200 250 300 350 400
Alc
oh
ol b
y V
olu
me
(AB
V)
Time (min)
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
D13
D14
D15
60
Figure 5.2 Ethanol Concentration vs Volume for Distillations D7 – D15
As you can see in Figure 5.2, not all distillations collected the same volume of distillate, therefore
to compare the data, the x-axis was normalized by dividing the volume at a given point by the
overall volume collected. This can be seen graphically in Figure 5.3 below.
Figure 5.3 Normalized Ethanol Concentration vs Volume for Distillations D7-D15
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 2 4 6 8 10 12 14 16
Alc
oh
ol b
y V
olu
me
(AB
V)
Volume (gals)
D7
D8
D9
D10
D11
D12
D13
D14
D15
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Alc
oh
ol b
y V
olu
me
Fraction of Volume Distilled
D7
D8
D9
D10
D11
D12
D13
D14
D15
61
5.1.2 Component Concentration Results
Figures 5.4 – 5.11 show the trend in compound concentration versus fraction of volume distilled
for distillation D7 – D15.
Figure 5.4 Normalized Acetaldehyde Concentration vs Volume for Distillations D7-D15
Figure 5.5 Normalized Acetone Concentration vs Volume for Distillations D7-D15
0
1
2
3
4
5
6
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Co
nce
ntr
atio
n (
g/L)
Fraction of Volume Distilled
D7
D8
D9
D10
D11
D12
D13
D14
D15
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Co
nce
ntr
atio
n (
g/L)
Fraction of Volume Distilled
D7
D8
D9
D10
D11
D12
D13
D14
D15
62
Figure 5.6 Normalized Ethyl Acetate Concentration vs Volume for Distillations D7-D15
Figure 5.7 Normalized Methanol Concentration vs Volume for Distillations D7-D15
0
5
10
15
20
25
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Co
nce
ntr
atio
n (
g/L)
Fraction of Volume Distilled
D7
D8
D9
D10
D11
D12
D13
D14
D15
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Co
nce
ntr
atio
n (
g/L)
Fraction of Volume Distilled
D7
D8
D9
D10
D11
D12
D13
D14
D15
63
Figure 5.8 Normalized Propanol Concentration vs Volume for Distillations D7-D15
Figure 5.9 Normalized Isobutanol Concentration vs Volume for Distillations D7-D15
0
0.05
0.1
0.15
0.2
0.25
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Co
nce
ntr
atio
n (
g/L)
Fraction of Volume Distilled
D7
D8
D9
D10
D11
D12
D13
D14
D15
0
0.2
0.4
0.6
0.8
1
1.2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Co
nce
ntr
atio
n (
g/L)
Fraction of Volume Distilled
D7
D8
D9
D10
D11
D12
D13
D14
D15
64
Figure 5.10 Normalized Butanol Concentration vs Volume for Distillations D7-D14
Figure 5.11 Normalized Isoamyl Alcohol Concentration vs Volume for Distillations D7-D15
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Co
nce
ntr
atio
n (
g/L)
Fraction of Volume Distilled
D7
D8
D9
D10
D11
D12
D13
D14
D15
0
1
2
3
4
5
6
7
8
9
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Co
nce
ntr
atio
n (
g/L)
Fraction of Volume Distilled
D7
D8
D9
D10
D11
D12
D13
D14
D15
65
5.1.3 Results by Type of Distillation
Overall, there were multiple types of distillations with and without reflux and with varying
initial feed to the pot. Cyclic versus conventional operation, as well as, low wines distillation
versus finishing run distillation. Below is a comparison table of operating parameters which led to
the separation of the distillations graphically into four different categories.
Table 5.2 Distillations Separated by Type
Operation
Type
Conventional
Operation
Conventional
Operation
Cyclic
Operation
Cyclic Operation
Distillation
Type
Low Wines Finishing Low Wines Finishing
Distillation
#
D7 D3,D11,D12 D8,D9,D10 D4,D5,D6,D13,D14,D15
Low Wines Distillation Results
Figure 5.12 below illustrates the low wines distillate results for cyclic versus conventional
operation.
Figure 5.12 Hydrometer ABV Reading vs Fraction of Volume Distilled for Low Wine
Distillations
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Hyd
rom
eter
AB
V
Fraction of Volume Distilled
D7 - Conventional
D8 - Cyclic (9,3)
D9 - Cyclic (6,3)
D10 - Cyclic (4,3)
66
Another important parameter to look at is the steam efficiency of each distillation. Steam efficiency
was calculated as the lbs. of steam used per proof gallon distilled. For low wines, this can be seen
in Figure 5.13 below.
Figure 5.13 Steam Efficiency for All Low Wine Distillations
The last measure of efficiency was percent recovery which was calculated by dividing the total
proof gallons distilled by the total initial proof gallons fed into the pot. For low wines this is
illustrated in Figure 5.14 below.
Figure 5.14 Percent Recovery for All Low Wine Distillations
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
lbs
Stea
m /
PG
dis
tille
d
D7 - Conventional
D8 - Cyclic (9,3)
D9 - Cyclic (6,3)
D10 - Cyclic (4,3)
0.80
0.82
0.84
0.86
0.88
0.90
0.92
0.94
0.96
0.98
1.00
1.02
% R
eco
vere
d D7 - Conventional
D8 - Cyclic (9,3)
D9 - Cyclic (6,3)
D10 - Cyclic (4,3)
67
Finishing Distillation Results
Figure 5.15 Hydrometer ABV Reading vs Fraction of Volume Distilled for Finishing
Distillations
Figure 5.16 Steam Efficiency for All Finishing Distillations
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Alc
oh
ol b
y V
olu
me
Fraction of Volume Distilled
D11 - Coventional
D12 - Conventional
D13 - Cyclic (6,3)
D14 - Cyclic (6,3)
D15 - Cyclic (6,3)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
lbs
Stea
m /
PG
dis
tille
d
D11 - Conventional
D12 - Conventional
D13 - Cyclic (6,3)
D14 - Cyclic (6,3)
D15 - Cyclic (6,3)
68
Figure 5.17 Percent Recovery for All Finishing Distillations
5.1.4 Results of Cuts Taken
Table 5.3 is a summary table of all the distillations in which cuts were taken. It shows the
cuts by volume, and the concentrations of the key components initially and in each cut.
Table 5.3 Cut Results for Distillations 11, 12, 13, 14 & 15
0.905
0.910
0.915
0.920
0.925
0.930
0.935
% R
eco
vere
dD11 - Conventional
D12 - Conventional
D13 - Cyclic (6,3)
D14 - Cyclic (6,3)
D15 - Cyclic (6,3)
Volume Acetaldehyde Acetone Methanol Ethanol Ethyl Acetate Propanol Isobutanol Butanol Isoamyl Alcohol ABV PG
gal g/L g/L g/L g/L g/L g/L g/L g/L g/L %
Initial Pot D11 36 0.03 0.00 0.06 151 0.13 0.03 0.09 0.03 0.50 0.19 13.82444
D12 36 0.03 0.00 0.07 215 0.13 0.04 0.13 0.04 0.70 0.27 19.60354
D13 36 0.02 0.00 0.07 222 0.15 0.04 0.13 0.05 0.77 0.28 20.25024
D14 33 0.02 0.00 0.07 214 0.12 0.04 0.13 0.04 0.72 0.27 17.86751
D15 33 0.01 0.00 0.07 241 0.13 0.05 0.15 0.05 0.84 0.31 20.18202
0
Heads D11 1.0 0.76 0.01 0.26 708 6.47 0.12 0.63 0.08 1.89 0.90 1.794442
D12 1.4 0.41 0.01 0.21 713 3.23 0.14 0.64 0.11 2.00 0.90 2.528523
D13 0.9 0.49 0.02 0.25 724 4.62 0.12 0.47 0.07 1.10 0.92 1.651692
D14 1.1 0.31 0.02 0.25 710 3.43 0.09 0.01 0.05 0.67 0.90 1.979305
D15 1.0 0.29 0.02 0.28 764 3.60 0.09 0.01 0.05 0.61 0.92 1.84
0
Hearts D11 5.3 0.03 0.00 0.22 709 0.10 0.14 0.52 0.14 3.35 0.90 9.530389
D12 7.7 0.02 0.00 0.16 599 0.11 0.14 0.49 0.17 3.12 0.76 11.68255
D13 9.1 0.02 0.00 0.19 694 0.12 0.15 0.49 0.17 3.02 0.88 16.01232
D14 8.2 0.00 0.00 0.18 663 0.04 0.15 0.48 0.17 2.91 0.84 13.776
D15 8.0 0 0 0.19 714 0 0.17 0.54 0.19 3.20 0.91 14.48422
0
Tails D11 2.1 0.00 0.00 0.19 288 0.00 0.02 0.01 0.02 0.12 0.37 1.533217
D12 5.8 0.00 0.00 0.13 293 0.00 0.04 0.03 0.03 0.21 0.37 4.308097
D13 1.8 0.00 0.00 0.15 250 0.00 0.02 0.01 0.02 0.07 0.32 1.142871
D14 1.6 0.00 0.00 0.14 215 0.00 0.02 0.02 0.01 0.05 0.27 0.864
D15 2.2 0 0 0.16 320 0.00 0.04 0.02 0.04 0.25 0.41 1.78342
69
To best compare these values, the volumes were converted into proof gallons, which as a reminder
is a gallon of alcohol at 50% alcohol by volume. The values were then normalized by dividing the
proof gallons of a given cut by the overall proof gallons fed into the still. This shows are more
direct comparisons of the amount of each cut produced as a percent of the initial volume. Figures
5.18 – 5.20 show quantitatively the normalized proof gallon of each cut in comparison by
distillation.
Figure 5.18 Normalized Proof Gallon of Heads Cut by Distillation
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
% P
G o
f In
itia
l PG
D11 - Conventional D12 - Conventional D13 - Cyclic (6,3)
D14 - Cyclic (6,3) D15 - Cyclic (6,3)
70
Figure 5.19 Normailized Proof Gallon of Hearts Cut by Distillation
Figure 5.20 Normailized Proof Gallon of Tails Cut by Distillation
5.1.5 Temperature Profile Results
Temperatures throughout the still and column were recorded during each distillation.
Figures 5.21-22 show conventional and cyclic temperature profiles for D11 and D 13.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
% P
G o
f In
itia
l PG
D11 - Conventional D12 - Conventional D13 - Cyclic (6,3)
D14 - Cyclic (6,3) D15 - Cyclic (6,3)
0
0.05
0.1
0.15
0.2
0.25
% P
G o
f In
itia
l PG
D11 - Conventional D12 - Conventional D13 - Cyclic (6,3)
D14 - Cyclic (6,3) D15 - Cyclic (6,3)
71
Figure 5.21 Temperature vs Time Profile for Conventional Operation (D11)
Figure 5.22 Temperature vs Time Profile for Cyclic Operation (D13)
5.2 Simulation Results
The previous modeling methods were applied to the system studied using MATLAB. The
MATLAB codes can be viewed in Appendix B. The initial conditions and other known parameters
used for the simulations can be seen below in Table 5.4.
0
10
20
30
40
50
60
70
80
90
100
0 20 40 60 80 100 120 140 160 180 200 220 240
Tem
per
atu
re (
⁰C)
Time (min)
T1 T2 T3 T4 T5 T6
0
10
20
30
40
50
60
70
80
90
100
0 20 40 60 80 100 120 140 160 180 200 220 240
Tem
per
atu
re (
⁰C)
Time (min)
T1 T2 T3 T4 T5 T6
72
Table 5.4 Parameters for Batch Distillation Simulation
Vapor Flow Rate (mole/min) 15
Reflux Ratio 1
Reboiler Hold Up (moles) 7000
Vapor Flow Period (min) 6
Total Operation Time (min) 185
Stage/Tray hold-up (moles) 56
Initial Condenser Mole fraction and
Temperature (°C) (X1,T1)
0.73,55
Initial Reboiler mole fraction and
Temperature (°C) (X2,T2)
0.64,55
Bottom tray mole fraction and Temperature
(°C) (X3,T3)
0.61,55
Middle tray mole fraction and Temperature
(°C) (X4,T4)
0.56,55
Top Tray mole fraction and Temperature (°C)
(X5,T5)
0.49,55
Partial condenser 2 mole fraction and
Temperature (°C) (X6,T6)
0.36,55
Initial Reboiler mole fraction and
Temperature (°C) (X7,T7)
0.10,55
For conventional operation, the partial differential and algebraic equations were set up in
the MATLAB function ‘Ethanol_Water_Conventional.’ initial values were entered into the global
code ‘EthanolWater_main’ which uses ODE15s to solve and plot the tray and still composition
and temperature over time. The resulting figures can be seen below in Figures 5.23 – 5.26.
73
Figure 5.23 MATLAB Simulation of Conventional Distillation: Composition vs Time
74
Figure 5.24 MATLAB Simulation of Conventional Distillation: Temperature vs Time
75
Figure 5.25 MATLAB Simulation of Conventional Distillation: Moles in Pot vs Time
76
Figure 5.26 MATLAB Simulation of Conventional Distillation: Moles in Distillate vs Time
For cyclic distillation modeling, the script ‘cyclic global’ was first ran to determine the state of the
system at the end of the vapor flow period. Plots of composition and temperature on the stages as
a function of time were generated for each vapor flow period. The results of the first vapor flow
period can be seen in Figures 5.27-28 below.
77
Figure 5.27 Stage and Reboiler Composition vs Time for First Vapor Flow Period
78
Figure 5.28 Stage and Reboiler Temperature vs Time for First Vapor Flow Period
The values obtained were then input into the script ‘liquid flow’ to determine the state of the system
at the end of the liquid flow period. The values obtained were then entered back into the script
‘cyclic global’ to determine the state of the system at the end of second vapor flow period. This
process continued for 25 cycles. The data was entered in an excel spreadsheet as it was collected.
The tables can be found in Appendix B. The graphical representation of the results can be seen in
Figures 5.29-30 below.
79
Figure 5.29 Cyclic Batch Distillation Tray and Still Composition vs Time for 25 Cycles
Figure 5.30 Cyclic Batch Distillation Tray and Still Temperature vs Time for 25 Cycles
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Eth
ano
l Mo
le F
rac.
Cycle
Total Con.
Part. Cond. 2
Top Tray
Middle Tray
Bottom Tray
Part. Cond. 1
Reboiler
50
55
60
65
70
75
80
85
90
95
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Tem
per
atu
re (
deg
C)
Cycle
Total Con.
Part. Cond. 2
Top Tray
Middle Tray
Bottom Tray
Part. Cond. 1
Reboiler
80
5.3 Reproducibility
Due to the length of each distillation to run as well as to analyze the samples, one
triplication experiment was run under the same conditions D13, D14 and D15 to prove
reproducibility. The normalized heads, hearts, and tails cuts were used to compare to the
normalized cuts taken in D11 and D12. First the heads, hearts and tails percentages were averaged.
Next, the standard deviation was calculated and used to determine a range of values which would
be accepted as within one standard deviation and therefore statistically the same. Then the values
from D11 and D12 were determined to be within or outside the respected ranges. The values can
be seen in Table 5.5 below.
Table 5.5 Statistical Analysis of Reproducibility
As seen in Table 5.4, the standard deviation of the triplicate is very small averaging around 0.017,
deeming the distillation results as the same and therefore reproducible. The distillation results of
D11 and D12 were determined to be statistically different from D13, D14 and D15 because the
values were all out of the range of 1 standard deviation.
D13 D14 D15 AVG STD Low End High End D11 D12
Heads 0.08 0.11 0.09 0.09 0.01 0.08 0.11 0.13 0.13
Hearts 0.79 0.77 0.72 0.76 0.03 0.73 0.79 0.69 0.60
Tails 0.06 0.05 0.09 0.06 0.01 0.05 0.08 0.11 0.22
81
CHAPTER 6: Discussion
6.1 Summary
In summary, cyclic distillation on the batch column used was able to show a decrease in
energy (steam) requirements for finishing runs but not in stripping runs. Cyclic distillation trends
included the following: ethanol concentration decreased at a slower rate compared to conventional
operation and as a result temperature profiles mimicked this phenomenon. It was shown that the
volume of hearts (product) was increased and volume of tails and heads (unwanted by-products)
was decreased. This research has shown that the application of cyclic distillation on spirit
production is a viable option for distilleries large and small. It is suggested that cyclic distillation
is applied to other types of spirits, to columns with a larger number of trays, and to trays with true
plug flow capability.
6.2 Conclusions
6.2.1 All Distillation Discussion
There are a few key differences when broadly comparing cyclic to conventional operation.
First, it was observed that for nearly the same initial volume and ABV as feed into the still, cyclic
operation took much longer to process than conventional operation. This trend makes sense
because during conventional operation, distillate is always being collected. During cyclic operation
distillate is only being collected during the vapor flow period, and during the beginning of the
liquid flow period when the condenser is draining.
Secondly, it was noticed that over time and over volume collected the hydrometer reading
stayed higher for longer. This is illustrated in Figure 5.3. Rather than start high and decrease
gradually throughout the distillation like in conventional operation, during cyclic operation the
82
distillate concentration stayed high and then dramatically decreased towards the end of the
distillation. Additionally, it was observed during each distillation that during conventional
operation the hydrometer reading started high, near 90% ABV, and then gradually decreased over
time. However, during cyclic operation, the hydrometer started high, near 90% ABV and gradually
decreased during the vapor flow period. However, it was also observed that in the beginning of a
vapor flow period, the hydrometer reading would increase about 3-5% from what it was at the end
of the liquid flow period. Both phenomena can be explained by the collected and essentially re-
distillation of the column and piping vapor and liquid contents. Rather than allowing the feed to
vaporize and condense and increasingly become less concentrated as the distillation continues,
some of the ethanol and other alcohols return to the pot and column, therefore causing a disruption
in this trend.
Lastly, Figures 5.21-22 display that the temperature in the column and top of the still
helmet increase at a slower rate during cyclic distillation compared to during conventional
operation. The temperature profiles were as expected because as the ethanol concentration
decreases, the water concentration increases. Since water has a higher boiling point than ethanol,
when there is a lower mole fraction of ethanol, the boiling point of the mixture is expected to
increase. As the ethanol concertation decreases at a slower rate throughout the pot and column, the
temperatures will increase at a slower rate as well.
6.2.2 Component Concentration Trends
Some compounds were not affected by cyclic operation including acetaldehyde, acetone,
and ethyl aetate. The trends can be seen in Figures 5.4 -5.6. This makes sense because of the
relative boiling points of these compounds compared to ethanol. The boiling point of these
83
compounds are all lower than ethanol, therefore they are concentrated in the distillate in the
beginning of the distillation and are not subject to cyclic operation.
Methanol also has a lower boiling point than ethanol (64.7°C) however it is well known in
industry and literature to start at a high concentration in the heads, decrease throughout the hearts
cut and then increase again in the tails cut. The trend observed in Figure 5.7 is like the trend Claus
and Berglund found in distilling a cherry distillate [43].
Higher alcohols including propanol, isobutanol, butanol and isoamyl alcohol with higher
boiling points than ethanol all exhibited similar trends and trend changes. A normal trend seen in
the previous report is a gradual increase from the beginning to the end of the distillation and a big
increase in the tails fraction [43]. This trend was inverted for the distillations when the reflux was
not engaged to the partial condensers. When the partial condensers are active, the vapor the heavier
and less volatile components of the vapor that travels through them are more likely to condense in
turn holding them back from entering the final condenser and into the distillate.
6.2.3 Low Wines vs Finishing Tends
Cyclic distillation of low wines did not seem to have any noticeable changes when
compared to conventional distillation of low wines. The efficiency measures compared in Figure
5.13 and 5.14 show conventional operation to be more efficient than cyclic operation. This is
attributed to the fact that the initial ABV of the feed is very low, roughly 10% ABV and that there
are only three trays which can allow cyclic distillation to occur. In distilled sprit products this is
one of the most energy intensive operations, because it takes more energy to heat a mixture that is
90% water than a mixture that is 70-75% water, and thus would be an ideal application of the
84
benefits of cyclic distillation. It is believed that with more trays and trays that allow true plug flow,
the energy saving advantage of cyclic distillation would be observed.
Increases in efficiency measures were seen in the finishing runs observed in Figures 5.15-
17. In cyclic distillations D13-D15 the ABV reading on the hydrometer remained higher for longer
compared to the identical distillation D11 for the reasons already explained in the ‘All Distillation
Discussion.’ Additionally, the steam usage per proof gallon distilled was the lower for D13-D15
compared to the identical conventional operation D11, proving an increasing in energy efficiency.
Lastly, the percent recovery values were too spread out to be able to confirm the percent recovery
was better in cyclic versus conventional operation.
6.2.4 Effects on Cuts
The last five distillations were used to identify what effect cyclic distillation had on the
heads, hearts and tails cuts. The results can be referred to in Table 5.3 and Figures 5.18 -20. The
heads cut was statistically smaller during cyclic operation than in conventional operation. The
hearts cut was larger for cyclic operation than in conventional operation and lastly the tails cut was
drastically smaller for cyclic operation than in conventional operation.
This is best explained by the earlier mentioned phenomenon that the ethanol concentration
stays higher for longer throughout the distillation during cyclic operation. If the ethanol
concentration remains high and the fusel oils/ congeners do not start coming through, the distillate
remains clean and pleasant to the distiller and the hearts cut increases in proof gals. This
consequence allows most of the ethanol to remain in the hearts cut. Since the proof gallons
collected in the hearts cut increased, the proof gallons in the tails cut decreased.
85
Based on the previous observations it can be concluded that cyclic distillation causes an
increase in proof gallon of hearts. This in turn increases the production efficiency and decreases
energy requirement per proof gal of product. In addition, decreasing the volume of by products
produced reduces the storage and processing requirements, decreasing cost per proof gal produced.
6.2.6 Reflux Observations and Discussion
It was observed that regardless of mode of operation, when reflux to the partial condenser
was turned on a smaller volume of hearts was collected, however it was higher in alcohol by
volume. This is because when the partial condensers are active, they act as an additional stage by
sending some condensed vapor to the preceding tray or to the reboiler/pot, depending on the
location of the partial condenser. When the partial condensers are not active, vapor travels past
this ‘stage.’ Overall, not having the partial condensers active causes less rectification in the column
therefore it is suitable to assume they will be used when making most types of spirits.
6.2.7 Simulation Discussion
Results from the simulation and modeling of conventional and cyclic operation can be seen
in Figures 5.23 -5.26 and Figures 5.27-5.30, respectively. The trends seen in the data were also
seen in the simulations. Temperatures in conventional operation increased at a faster rate than in
cyclic operation. Distillate ethanol concentration decreased over time at a faster rate than in cyclic
operation. During cyclic distillation modeling, some of the ethanol is transferred to the tray below,
causing the ethanol to be held back in the column and thus the tray composition to increase and
decrease slightly from cycle to cycle. This trend of cyclic tray composition is seen graphically in
Figure 5.29. The model calculates the temperature based on the ethanol composition, therefore
temperature and concentration are directly proportional and can be seen in Figure 5.30.
86
Lastly, given the same processing time cyclic operation was able to process less moles in
the pot than in conventional operation. This was also seen in the data and can be explained on the
basis that during the liquid flow period, no distillate is collected, leading to a longer distillation
time.
6.3 Limitations and Sources of Error
One key limitation is the lack of separation between liquids drained from tray to tray. The
trays in the Carl still have small holes that allow the liquid to fall to the previous tray regardless of
whether it is during the vapor flow or the liquid flow period. Therefore, it was not possible to
separate the liquid hold up traveling from tray to tray during each liquid flow period. Since the still
used was not specifically designed for cyclic distillation, the still was not able to achieve the 200%
efficiency increase predicted by theoretical models.
Is it important to note that cuts are taken by the distiller and are predominantly made by
sensory analysis. In industry, it is common for distillers to go through specific sensory training to
become more attune with the flavor and aroma of the compounds in the spirit. The distiller of this
research has two-year part-time experience distilling which is more than the average researcher,
but less than a full-time trained distiller. Many factors play a role in sensory analysis including
what and what time the distiller has last ate or drank can affect the sensory perception, therefore
cuts cannot be 100% consistent.
Another important note is the inconsistency of initial volume and proof gallons in the feed.
When large volume and large equipment is used, it becomes increasingly hard to duplicate an exact
volume and initial ABV due to the necessary pumping into the still, and losses associated with the
hoses.
87
6.4 Recommendations for Future Research and Application
It is strongly suggested that cyclic distillation be applied to other types of spirits. This
research was solely based on brandy made from fermented apple cider, both the stripping and the
finishing runs. While the energy requirements are mostly likely the same, different spirits are
fermented with different yeast strains and ultimately have very different components and
concentrations of key flavor and aroma attributes. Rum, whiskey, gin and vodka production are all
sprits that can be further investigated. Legally, vodka must be distilled to 95% alcohol by volume,
so its production requires more trays than all other spirits. Therefore, the application of cyclic
distillation for vodka production is a promising addition to this research for both energy
conservation and efficiency purposes. In addition to other spirits, implementing the already
designed Maleta© trays or a new tray design which maximizes true plug flow would increase the
positive effects of cyclic distillation.
88
APPENDICES
89
APPENDIX A: Graphical Distillation Results
Tray by Tray Results:
Distillation
Figure A.1 Normal Operation 1, Tray 1 Graphical Results
Figure A.2 Normal Operation 1, Tray 2 Graphical Results
0
100
200
300
400
500
600
700
0
1
2
3
4
5
6
35 54 72 104 119 137 154
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
0
1
2
3
4
5
6
1 2 3 4 5 6 7Et
han
ol C
on
cen
trat
ion
(g/
L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
90
Figure A.3 Normal Operation 1, Tray 3 Graphical Results
Distillation 2
Figure A.4 Normal Operation 2 , Tray 1 Graphical Results
0
100
200
300
400
500
600
700
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
1 2 3 4 5 6 7
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
50
100
150
200
250
300
0
0.5
1
1.5
2
2.5
3
3.5
4
67 78 89 103 117 135 155 168 183 200 213 229 248 266 284Et
han
ol C
on
cen
trat
ion
(g/
L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
91
Figure A.5 Normal Operation 2 , Tray 2 Graphical Results
Figure A.6 Normal Operation 2 , Tray 3 Graphical Results
0
50
100
150
200
250
300
350
400
450
0
1
2
3
4
5
6
7
68 79 90 103 118 135 155 169 183 201 213 229 248 267 285
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
50
100
150
200
250
300
350
400
450
500
0
1
2
3
4
5
6
7
8
69 80 90 104 118 137 156 170 184 202 214 250 268 286Et
han
ol C
on
cen
trat
ion
(g/
L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
92
Distillation 3
Figure A.7 Normal Operation 3 , Tray 1 Graphical Results
Figure A.8 Normal Operation 3 , Tray 2 Graphical Results
0
100
200
300
400
500
600
0
0.5
1
1.5
2
2.5
3
3.5
4
55 65 77 86 96 107 116 126 137 146 157 168 180 191 202
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
0
1
2
3
4
5
6
55 66 77 86 97 108 117 127 137 147 157 169 181 191 202
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
93
Figure A.9 Normal Operation 3 , Tray 3 Graphical Results
Distillation 4
Figure A.10 Cyclic Distillation 1 (Vapor 9 min, Liquid 3 min) , Tray 1 Graphical Results
0
100
200
300
400
500
600
0
1
2
3
4
5
6
57 66 78 87 97 108 118 127 138 147 158 169 182 192 203
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
0
100
200
300
400
500
600
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
31 44 56 67 80 92 104 116 130 142 155 167 178 192 204 216 230Et
han
ol C
on
cen
trat
ion
(g/L
)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
94
Figure A.11 Cyclic Distillation 1 (Vapor 9 min, Liquid 3 min) , Tray 2 Graphical Results
Figure A.12 Cyclic Distillation 1 (Vapor 9 min, Liquid 3 min) , Tray 3 Graphical Results
0
100
200
300
400
500
600
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
32 45 56 67 80 93 104 116 130 142 155 168 179 192 205 217 231
Eth
ano
l Co
nce
ntr
atio
n(g
/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
0
100
200
300
400
500
600
0
1
2
3
4
5
6
33 45 57 68 81 93 104 116 130 143 156 168 179 192 205 217 231
Eth
ano
l Co
nce
ntr
atio
n(g
/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
95
Distillation 5
Figure A.13 Cyclic Distillation 2 (Vapor 6 min, Liquid 3 min) , Tray 1 Graphical Results
Figure A.14 Cyclic Distillation 2 (Vapor 6 min, Liquid 3 min) , Tray 2 Graphical Results
0
100
200
300
400
500
600
700
0
1
2
3
4
5
6
72
6
55
64
73
82
92
10
1
11
0
12
0
13
1
13
9
14
8
15
8
16
8
17
87
18
6
19
6
20
5
21
4
22
4
23
3
24
2
25
1
26
1
27
0
28
0
29
0
29
9
30
9
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
0
100
200
300
400
500
600
700
0
1
2
3
4
5
6
7
8
27 55 65 75 84 93 102111122132141149159168178186196206215224233242252261270281290299
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
96
Figure A.15 Cyclic Distillation 2 (Vapor 6 min, Liquid 3 min) , Tray 3 Graphical Results
Distillation 6
Figure A.16 Cyclic Distillation 3 (Vapor 4 min, Liquid 3 min) , Tray 1 Graphical Results
0
100
200
300
400
500
600
700
800
0
1
2
3
4
5
6
28
56
66
75
84
94
10
3
11
2
12
3
13
3
14
1
15
0
16
0
16
9
17
8
18
7
19
7
20
6
21
5
22
4
23
3
24
3
25
2
26
1
27
1
28
1
29
1
30
0
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
0
1
2
3
4
5
6
7
8
39 59 73 88 103 117 132 147 161 177 191 206 221 236 250 268 282 298 313 327 342
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
97
Figure A.17 Cyclic Distillation 3 (Vapor 4 min, Liquid 3 min) , Tray 2 Graphical Results
Figure A.18 Cyclic Distillation 3 (Vapor 4 min, Liquid 3 min) , Tray 3 Graphical Results
0
100
200
300
400
500
600
700
0
1
2
3
4
5
6
7
8
9
10
40 59 74 88 103 117 131 148 162 177 191 206 221 236 251 268 283 298 313 327 343
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
800
0
1
2
3
4
5
6
7
8
40 60 74 89 103 118 132 148 162 178 191 207 222 236 251 268 283 298 313 328 343
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
98
Distillation 7
Figure A.19 Normal Operation – Low Wines , Tray 1 Graphical Results
Figure A.20 Normal Operation – Low Wines , Tray 2 Graphical Results
0
10
20
30
40
50
60
70
80
90
0
0.05
0.1
0.15
0.2
0.25
30 38 46 53 60 67 70
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
50
100
150
200
250
300
0
0.5
1
1.5
2
2.5
3
3.5
31 38 46 53 61 68 70
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
99
Figure A.21 Normal Operation – Low Wines , Tray 3 Graphical Results
Distillation 8
Figure A.22 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 9 min, Liquid 3 min)
Tray 1 Graphical Results
0
50
100
150
200
250
300
350
0
1
2
3
4
5
6
31 39 46 53 61 68 71
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
50
100
150
200
250
0
0.5
1
1.5
2
2.5
3
43 53 76 89 104 115 126 138
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
100
Figure A.23 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 9 min, Liquid 3 min)
Tray 2 Graphical Results
Figure A.24 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 9 min, Liquid 3 min)
Tray 3 Graphical Results
0
100
200
300
400
500
600
0
1
2
3
4
5
6
7
8
9
10
43 54 77 90 104 115 126 139
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
50
100
150
200
250
300
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
43 54 77 90 104 116 127 139
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
101
Distillation 9
Figure A.25 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 6 min, Liquid 3 min)
Tray 1 Graphical Results
Figure A.26 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 6 min, Liquid 3 min)
Tray 2 Graphical Results
0
50
100
150
200
250
0
0.5
1
1.5
2
2.5
40 52 60 71 78 89 99 107 116 125 134 143
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
50
100
150
200
250
300
350
0
0.5
1
1.5
2
2.5
3
40 53 61 71 79 90 99 107 117 125 134 143
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
102
Figure A.27 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 6 min, Liquid 3 min)
Tray 3 Graphical Results
Distillation 10
Figure A.28 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 4 min, Liquid 3 min)
Tray 1 Graphical Results
0
50
100
150
200
250
300
350
0
1
2
3
4
5
6
41 53 61 72 79 90 99 108 117 126 135 144
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
50
100
150
200
250
300
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
64 74 81 88 96 103 111 119 126 134 141 149 156 163 171 178 186 193
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate Methanol Propanol
Isobutanol Butanol Isoamyl Alcohol Ethanol
103
Figure A.29 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 4 min, Liquid 3 min)
Tray 2 Graphical Results
Figure A.30 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 4 min, Liquid 3 min)
Tray 3 Graphical Results
0
50
100
150
200
250
300
350
0
1
2
3
4
5
6
7
64 74 82 88 96 103 111 119 126 134 142 149 156 164 172 179 186 193
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate Methanol Propanol
Isobutanol Butanol Isoamyl Alcohol Ethanol
0
50
100
150
200
250
300
350
400
0
1
2
3
4
5
6
65 75 82 89 97 105 112 120 127 135 143 150 157 165 172 179 186 194
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
104
Distillation 11
Figure A.31 Normal Operation – Brandy Finishing Run w/ reflux - Tray 1 Graphical
Results
Figure A.32 Normal Operation – Brandy Finishing Run w/ reflux - Tray 2 Graphical
Results
0
50
100
150
200
250
300
350
400
450
500
0
1
2
3
4
5
6
7
8
9
31 40 50 59 66 74 83 93 101 109 116 126
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time(min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
0
2
4
6
8
10
12
32 41 50 59 67 75 84 93 101 109 117 126
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time(min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
105
Figure A.33 Normal Operation – Brandy Finishing Run w/ reflux - Tray 3 Graphical
Results
Distillation 12
Figure A.34 Normal Operation – Brandy Finishing Run w/o reflux - Tray 1 Graphical
Results
0
100
200
300
400
500
600
700
0
2
4
6
8
10
12
32 41 49 59 67 75 84 93 102 109 117 128
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time(min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
50
100
150
200
250
300
350
400
450
500
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
36 43 51 57 65 72 79 85 92 99 106 114 120 127 134 141 148 155
Tim
e (m
in)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
106
Figure A.35 Normal Operation – Brandy Finishing Run w/o reflux - Tray 2 Graphical
Results
Figure A.36 Normal Operation – Brandy Finishing Run w/o reflux - Tray 3 Graphical
Results
0
100
200
300
400
500
600
0
1
2
3
4
5
6
36 44 51 58 65 72 79 86 93 99 107 114 121 128 135 142 148 155
Tim
e (m
in)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
0
1
2
3
4
5
6
37 44 51 58 65 72 79 86 93 100 107 114 121 128 135 142 149 155
Tim
e (m
in)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
107
Distillation 13
Figure A.37 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 1
Graphical Results
Figure A.38 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 2
Graphical Results
0
100
200
300
400
500
600
700
0
1
2
3
4
5
6
7
8
37 48 57 64 73 80 88 95 103109118124133140148155163170177184192200207215 Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
800
0
2
4
6
8
10
12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
108
Figure A.39 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 3
Graphical Results
Distillation 14
Figure A.40 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 1
Graphical Results
0
100
200
300
400
500
600
700
800
0
2
4
6
8
10
12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
0
2
4
6
8
10
12
34 43 50 56 63 71 78 85 92 100 107 114 122 129 136 143 148 158 Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
109
Figure A.41 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 2
Graphical Results
Figure A.42 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 3
Graphical Results
0
100
200
300
400
500
600
700
0
2
4
6
8
10
12
14
34 43 50 56 64 71 78 86 92 100 107 115 122 129 136 143 150 158 Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
0
2
4
6
8
10
12
14
34 43 50 56 64 71 78 86 93 100 108 115 122 129 137 144 151 158 Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
110
Distillation 15
Figure A.43 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 1
Graphical Results
Figure A.44 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 2
Graphical Results
0
100
200
300
400
500
600
0
1
2
3
4
5
6
7
8
9
28 38 45 52 59 67 74 81 89 96 104 111 118 125 134 141 149 157 Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
0
2
4
6
8
10
12
28 38 46 52 59 67 74 81 89 96 104 112 118 126 134 142 149 157 164 Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
111
Figure A.45 Cyclic Operation – Brandy Finishing Tray by Tray w/ Reflux - Tray 3
Graphical Results
0
100
200
300
400
500
600
700
0
2
4
6
8
10
12
29 38 46 52 59 67 75 82 89 96 105 112 118 126 135 142 149 157 164
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
112
Distillate Results
Distillation 1
Figure A.46 Normal Operation 1 – Distillate Graphical Results
Distillation 2
Figure A.47 Normal Operation 2 – Distillate Graphical Results
480
500
520
540
560
580
600
620
640
660
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
1 2 3 4 5 6 7
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
62 77 88 101 116 133 152 166 180 197 211 226 245 264 282
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
113
Distillation 3
Figure A.48 Normal Operation 3 – Distillate Graphical Results
Distillation 4
Figure A.49 Cyclic Distillation 1 (Vapor 9 min, Liquid 3 min) – Distillate Graphical
Results
0
100
200
300
400
500
600
700
0
0.5
1
1.5
2
2.5
3
3.5
4
56 67 78 87 97 109 118 127 138 148 158 169 182 193 203 217 230 246
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
31 43 57 68 81 94 105 117 131 143 156 169 180 193 206 218 232 242 255 269
Eth
ano
l Co
nce
ntr
atio
n(g
/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
114
Distillation 5
Figure A.50 Cyclic Distillation 1 (Vapor 6 min, Liquid 3 min) – Distillate Graphical
Results
Distillation 6
Figure A.51 Cyclic Distillation 1 (Vapor 4 min, Liquid 3 min) – Distillate Graphical
Results
0
100
200
300
400
500
600
700
800
0
1
2
3
4
5
62
95
66
57
48
49
31
02
11
21
22
13
21
41
15
01
59
16
91
79
18
71
97
20
62
15
22
82
34
24
32
53
26
22
71
28
22
91
30
03
09
31
83
27
33
73
46
35
4
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
800
0
1
2
3
4
5
6
38 59 73 87 102 116 132 147 161 176 190 206 220 236 252 267 282 297 312 326 342
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate Methanol Propanol
Isobutanol Butanol Isoamyl Alcohol Ethanol
115
Distillation 7
Figure A.52 Normal Operation – Low Wines – Distillate Graphical Results
Distillation 8
Figure A.53 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 9 min, Liquid 3 min)
Distillate Graphical Results
0
100
200
300
400
500
600
700
800
0
1
2
3
4
5
6
7
8
29 37 44 52 59 67 73
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
800
0
2
4
6
8
10
12
42 52 76 87 100 112 125 137
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
116
Distillation 9
Figure A.54 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 6 min, Liquid 3 min)
Distillate Graphical Results
Distillation 10
Figure A.55 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 4 min, Liquid 3 min)
Distillate Graphical Results
0
100
200
300
400
500
600
700
800
900
0
2
4
6
8
10
12
39 52 60 70 78 87 97 106 116 125 133 142
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
800
0
2
4
6
8
10
12
14
16
63 74 81 87 96 103 111 118 125 133 141 148 156 163 171 178 185 192
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
117
Distillation 11
Figure A.56 Normal Operation – Brandy Finishing Run w/o reflux - Distillate Graphical
Results
Distillation 12
Figure A.57 Normal Operation – Brandy Finishing Run w/ reflux - Distillate Graphical
Results
0
100
200
300
400
500
600
700
800
0
5
10
15
20
25
31 40 50 57 65 74 83 92 100 108 116 126
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time(min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
800
0
1
2
3
4
5
6
35 43 50 57 64 71 78 85 92 99 106 113 120 127 134 141 148 155
Tim
e (m
in)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
118
Distillation 13
Figure A.58 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Distillate
Graphical Results
Distillation 14
Figure A.59 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Distillate
Graphical Results
0
100
200
300
400
500
600
700
800
0
2
4
6
8
10
12
14
36 47 56 64 72 80 87 95 102 109 118 124 133 140 147 155 162 169 177 184 191 199 207 215
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
800
900
0
2
4
6
8
10
12
14
16
33 42 49 56 63 70 77 85 92 99 107 114 121 128 135 143 150 157 164
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
119
Distillation 15
Figure A.60 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Distillate
Graphical Results
0
100
200
300
400
500
600
700
800
0
2
4
6
8
10
12
28 37 45 51 58 66 74 80 88 95 104 111 118 125 134 141 149 156 163
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
120
Bottom Column and Cycled Volume Results
Distillation 1
Figure A.61 Normal Operation 1 – Bottom Column Graphical Results
Distillation 2
Figure A.62 Normal Operation 2 – Bottom Column Graphical Results
0
100
200
300
400
500
600
0
1
2
3
4
5
6
53 71 103 118 136 153 Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
50
100
150
200
250
300
350
400
450
500
0
0.5
1
1.5
2
2.5
3
3.5
67 78 89 102 117 133 154 167 182 199 212 228 247 265 283Et
han
ol C
on
cen
trat
ion
(g/
L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
121
Distillation 3
Figure A.63 Normal Operation 3 – Bottom Column Graphical Results
Distillation 4
Figure A.64 Cyclic Distillation 1 (Vapor 9 min, Liquid 3 min) – Cycled Volume Graphical
Results
0
100
200
300
400
500
600
700
0
0.5
1
1.5
2
2.5
3
3.5
4
55 65 76 85 95 107 115 125 136 146 156 167 180 190 201
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Eth
ano
l Co
nce
ntr
atio
n(g
/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
122
Distillation 5
Figure A.65 Cyclic Distillation 1 (Vapor 6 min, Liquid 3 min) – Cycled Volume Graphical
Results
Distillation 6
Figure A.66 Cyclic Distillation 1 (Vapor 4 min, Liquid 3 min) – Cycled Volume Graphical
Results
0
100
200
300
400
500
600
700
0
1
2
3
4
5
65
0
60
70
79
88
97
10
7
11
6
12
6
13
5
14
4
15
4
16
4
17
3
18
2
19
1
20
1
21
0
21
9
22
8
23
7
24
6
25
6
26
5
27
6
28
5
29
4
30
4
31
3
32
2
33
1
34
1
35
0
35
9
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
0
1
2
3
4
5
6
7
54 69 84 99 114 129 144 159 174 188 202 217 232 257 272 287 301 316 331 346 360
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
123
Distillation 7
Figure A.67 Normal Operation – Low Wines – Bottom Column Graphical Results
Distillation 8
Figure A.68 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 9 min, Liquid 3 min)
Cycled Volume Graphical Results
0
50
100
150
200
250
300
350
400
450
00.20.40.60.8
11.21.41.61.8
2
30 37 45 52 60 67 70
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
10
20
30
40
50
60
70
80
90
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
48 71 83 96 109 120 133 145Et
han
ol C
on
cen
trat
ion
(g/
L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
124
Distillation 9
Figure A.69 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 6 min, Liquid 3 min)
Cycled Volume Graphical Results
Distillation 10
Figure A.70 Cyclic Operation – Low Wines, Cyclic Distillation (Vapor 4 min, Liquid 3 min)
Cycled Volume Graphical Results
0
20
40
60
80
100
120
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
45 55 65 74 84 94 103 112 121 130 139 148
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
0
10
20
30
40
50
60
70
80
90
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
68 77 81 87 96 103 111 118 125 133 141 148 160 163 171 178 189 196
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)Acetaldehyde Acetone Ethyl AcetateMethanol Propanol IsobutanolButanol Isoamyl Alcohol Ethanol
125
Distillation 11
Figure A.71 Normal Operation – Brandy Finishing Run w/o reflux - Bottom Column
Graphical Results
Distillation 12
Figure A.72 Normal Operation – Brandy Finishing Run w/ reflux - Bottom Column
Graphical Results
0
100
200
300
400
500
600
0
1
2
3
4
5
6
33 42 52 60 67 76 85 94 101 110 117 127
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time(min)Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
0
0.5
1
1.5
2
2.5
3
3.5
4
37 45 52 58 66 73 79 86 93 100 108 115 122 128 135 142 149 156
Tim
e (m
in)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
126
Distillation 13
Figure A.73 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Cycled Volume
Graphical Results
Distillation 14
Figure A.74 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Cycled Volume
Graphical Results
0
100
200
300
400
500
600
700
800
0
2
4
6
8
10
12
43 52 61 69 76 84 90 99 106 114 121 129 137 144 150 159 166 173 181 189 196 204 212
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
0
100
200
300
400
500
600
700
0
1
2
3
4
5
6
7
8
9
40 46 53 60 68 76 83 89 96 104 111 119 126 133 141 148 155 163
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
127
Distillation 15
Figure A.75 Cyclic Operation – Tray by Tray Finishing Run w/ reflux – Cycled Volume
Graphical Results
0
100
200
300
400
500
600
0
1
2
3
4
5
6
7
8
9
10
34 42 49 56 63 70 79 86 92 100 107 115 122 130 139 146 153 161
Eth
ano
l Co
nce
ntr
atio
n (
g/L)
Co
nce
ntr
atio
n (
g/L)
Time (min)
Acetaldehyde Acetone Ethyl Acetate
Methanol Propanol Isobutanol
Butanol Isoamyl Alcohol Ethanol
128
Component Results
Figure A.76 Acetaldehyde Concentration vs Time for All Distillations
Figure A.77 Acetone Concentration vs Time for All Distillations
0.000
1.000
2.000
3.000
4.000
5.000
6.000
0 50 100 150 200 250 300 350 400
Co
nce
ntr
atio
n (
g/L)
Time (min)
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0 50 100 150 200 250 300 350 400
Co
nce
ntr
atio
n (
g/L(
Time(min)
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
D13
D14
D15
129
Figure A.78 Ethyl Acetate Concentration vs Time for All Distillations
Figure A.79 Methanol Concentration vs Time for All Distillations
0.000
5.000
10.000
15.000
20.000
25.000
0 50 100 150 200 250 300 350 400
Co
nce
ntr
atio
n (
g/L)
Time (min)
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
D13
D14
D15
0.000
0.050
0.100
0.150
0.200
0.250
0.300
0.350
0.400
0.450
0 50 100 150 200 250 300 350 400
Co
nce
ntr
atio
n (
g/L)
Time (min)
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
D13
D14
D15
130
Figure A.80 Ethanol Concentration vs Time for All Distillations
Figure A.81 Propanol Concentration vs Time for All Distillations
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 50 100 150 200 250 300 350 400
Hyd
rom
eter
AB
V
Time (min)
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
D13
D14
0.000
0.050
0.100
0.150
0.200
0.250
0.300
0 50 100 150 200 250 300 350 400
Co
nce
ntr
atio
n (
g/L)
Time (min)
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
131
Figure A.82 Isobutanol Concentration vs Time for All Distillations
Figure A.83 Butanol Concentration vs Time for All Distillations
0.000
0.200
0.400
0.600
0.800
1.000
1.200
0 50 100 150 200 250 300 350 400
Co
nce
ntr
atio
n (
g/L)
Time (min)
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
D13
D14
D15
0.000
0.050
0.100
0.150
0.200
0.250
0.300
0.350
0.400
0.450
0 50 100 150 200 250 300 350 400
Co
nce
ntr
atio
n (
g/L)
Time (min)
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
D13
D14
D15
132
Figure A.84 Isoamyl Alcohol Concentration vs Time for All Distillations
-1.000
0.000
1.000
2.000
3.000
4.000
5.000
6.000
7.000
8.000
9.000
0 50 100 150 200 250 300 350 400
Co
nce
ntr
atio
n (
g/L)
Time (min)
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
D13
D14
D15
133
APPENDIX B: MATLAB Codes
Conventional Simulation MATLAB Codes
134
135
136
137
Cyclic Simulation MATLAB Codes
138
139
140
141
MATLAB Cyclic Data
Table B.1 Cyclic Distillation Simulation Data
Cycle Period Reboiler
Hold Up X1 T1 X2 T2 X3 T3 X4 T4 X5 T5 X6 T6 X7 T7
Initial - 7000 0.7342 55 0.6417 55 0.6086 55 0.5623 55 0.4942 55 0.3625 55 0.0995 55
VFP 6798 0.7898 74.1203 0.6308 73.7323 0.5902 73.7326 0.5137 73.8453 0.3238 75.204 0.1261 81.694 0.0943 84.194 #VALUE!
LFP 6854 0.7898 74.1203 0.7898 73.7323 0.7499 73.7326 0.7027 73.8453 0.6269 75.204 0.0946 81.694 0.0946 84.194 29.194
VFP 6764 0.8252 74.3334 0.775 74.0509 0.7269 73.8816 0.5966 73.7298 0.225 77.236 0.0907 84.529 0.0893 84.6559 0.4619
LFP 6820 0.8252 74.3334 0.8252 74.0509 0.8055 73.8816 0.7637 73.7298 0.656 77.236 0.0893 84.529 0.0892 84.6559 0.4619
VFP 6730 0.8401 74.4487 0.814 74.2591 0.7797 74.073 0.647 73.7435 0.2271 77.171 0.0854 85.0367 0.0841 85.1597 0.5038
LFP 6786 0.8401 74.4487 0.8401 74.2591 0.8281 74.073 0.7919 73.7435 0.6789 77.171 0.0842 85.0367 0.0842 85.1597 0.5038
VFP 6696 0.8472 74.5093 0.8306 74.3744 0.8012 74.1828 0.6702 73.7688 0.225 77.2287 0.0802 85.563 0.0791 85.683 0.5233
LFP 6752 0.8472 74.5093 0.8472 74.3744 0.838 74.1828 0.8044 73.7688 0.6885 77.2287 0.0791 85.563 0.0791 85.683 0.5233
VFP 6662 0.8507 74.5408 0.8383 74.434 0.8102 74.236 0.6766 73.7781 0.2145 77.5449 0.0751 86.1087 0.0741 86.2252 0.5422
LFP 6718 0.8507 74.5408 0.8507 74.434 0.8426 74.236 0.8094 73.7781 0.6904 77.5449 0.0741 86.1087 0.0741 86.2252 0.5422
VFP 6628 0.8524 74.557 0.8418 74.4636 0.8137 74.4636 0.6744 73.775 0.1988 78.0698 0.0701 86.673 0.0692 86.7857 0.5605
LFP 6684 0.8524 74.557 0.8524 74.4636 0.8447 74.4636 0.8106 73.775 0.6883 78.0698 0.0692 86.673 0.0692 86.7857 0.5605
VFP 6594 0.8533 74.565 0.8435 74.4774 0.8147 74.2648 0.667 73.7651 0.1808 78.7612 0.0653 87.2548 0.0644 87.3634 0.5777
LFP 6650 0.8533 74.565 0.8533 74.4774 0.8456 74.2648 0.8099 73.7651 0.684 78.7612 0.0644 87.2548 0.0644 87.3634 0.5777
VFP 6560 0.8537 74.5688 0.8441 74.4826 0.8143 74.2621 0.6564 73.7529 0.1617 79.6105 0.0605 87.8528 0.0597 87.9568 0.5934
LFP 6616 0.8537 74.5688 0.8537 74.4826 0.8458 74.2621 0.8079 73.7529 0.6787 79.6105 0.0597 87.8528 0.0597 87.9568 0.5934
VFP 6526 0.8538 74.5701 0.8441 74.4832 0.813 74.2532 0.643 73.741 0.1425 80.6161 0.0559 88.4649 0.0552 88.564 0.6072
LFP 6582 0.8538 74.5701 0.8538 74.4832 0.8457 74.2532 0.8051 73.741 0.6727 80.6161 0.0552 88.4649 0.0552 88.564 0.6072
VFP 6492 0.8538 74.5702 0.8439 74.4812 0.8112 74.2431 0.627 73.7319 0.1236 81.783 0.0515 89.0888 0.0509 89.1828 0.6188
LFP 6548 0.8538 74.5702 0.8538 74.4812 0.8452 74.2431 0.8017 73.7319 0.6661 81.783 0.0509 89.0888 0.0509 89.1828 0.6188
VFP 6458 0.8538 74.5699 0.8535 74.4782 0.8092 74.2307 0.6086 73.728 0.1054 83.1273 0.0473 89.7218 0.0467 89.8106 0.6278
LFP 6514 0.8538 74.5699 0.8538 74.4782 0.8499 74.2307 0.7976 73.728 0.6592 83.1273 0.0467 89.7218 0.0467 89.8106 0.6278
VFP 6424 0.8537 74.569 0.8431 74.4741 0.8067 74.2162 0.5875 73.7332 0.0883 84.6185 0.0432 90.3607 0.0427 90.4441 0.6335
LFP 6480 0.8537 74.569 0.8537 74.4741 0.8443 74.2162 0.793 73.7332 0.652 84.6185 0.0427 90.3607 0.0427 90.4441 0.6335
VFP 6390 0.8535 74.5676 0.8423 74.4681 0.8035 74.1975 0.5631 73.7515 0.0738 86.1187 0.0393 91.0016 0.0393 91.0795 0.6354
LFP 6446 0.8535 74.5676 0.8535 74.4681 0.8435 74.1975 0.7875 73.7515 0.6447 86.1187 0.0388 91.0016 0.0388 91.0795 0.6354
VFP 6356 0.8534 74.566 0.8416 74.462 0.8001 74.1784 0.5358 73.7905 0.0607 87.6971 0.0356 91.6406 0.0352 91.713 0.6335
LFP 6412 0.8534 74.566 0.8534 74.462 0.8427 74.1784 0.7814 73.7905 0.6372 87.6971 0.0352 91.6406 0.0252 91.713 0.6335
VFP 6322 0.8532 74.5646 0.8409 74.4563 0.7967 74.1594 0.5061 73.8574 0.0499 89.3456 0.0321 92.2738 0.0318 92.3407 0.6277
LFP 6378 0.8532 74.5646 0.8532 74.4563 0.8419 74.1594 0.7747 73.8574 0.6296 89.3456 0.0317 92.2738 0.0317 92.3407 0.6277
VFP 6288 0.8531 74.563 0.8401 74.4494 0.7923 74.1367 0.473 73.9666 0.0402 90.7527 0.0288 92.9051 0.0285 92.9666 0.6259
LFP 6344 0.8531 74.563 0.8531 74.4494 0.8409 74.1367 0.7673 73.9666 0.6219 90.7527 0.0285 92.9051 0.0285 92.9666 0.6259
VFP 6254 0.8529 74.5611 0.8392 74.4421 0.7876 74.113 0.4385 74.127 0.0329 92.0495 0.0258 93.5124 0.0255 93.5685 0.6019
LFP 6310 0.8529 74.5611 0.8529 74.4421 0.8398 74.113 0.7595 74.127 0.6142 92.0495 0.0255 93.5124 0.0255 93.5685 0.6019
VFP 6220 0.8526 74.5588 0.8381 74.4332 0.782 74.0858 0.4028 74.3536 0.0276 93.0784 0.023 94.0998 0.0227 94.1509 0.5824
LFP 6276 0.8526 74.5588 0.8526 74.4332 0.8385 74.0858 0.7514 74.3536 0.6066 93.0784 0.0227 94.0998 0.0227 94.1509
VFP 6186 0.8524 74.5568 0.8371 74.4253 0.7763 74.0601 0.3665 74.664 0.0232 94.0043 0.0204 94.6639 0.0201 94.7101
LFP 6242 0.8524 74.5568 0.8524 74.4253 0.8372 74.0601 0.743 74.664 0.5991 94.0043 0.0202 94.6639 0.0202 94.7101
VFP 6152 0.8522 74.5568 0.836 74.4253 0.7698 74.0601 0.3309 74.664 0.0197 94.0043 0.018 94.6639 0.0178 94.7101
LFP 6208 0.8522 74.5568 0.8522 74.4253 0.8357 74.0601 0.7347 74.664 0.5917 94.0043 0.0178 94.6639 0.0178 94.7101
VFP 6118 0.8519 74.5523 0.8349 74.4078 0.7627 74.0037 0.2964 75.5731 0.0169 95.4286 0.0158 95.707 0.0156 95.7442
LFP 6174 0.8519 74.5523 0.8519 74.4078 0.8341 74.0037 0.7265 75.5731 0.5846 95.4286 0.0156 95.707 0.0156 95.7442
VFP 6084 0.8517 74.55 0.8337 74.3988 0.7552 73.9756 0.2635 76.1954 0.0145 95.9995 0.0138 96.1806 0.0137 96.2137
LFP 6140 0.8517 74.55 0.8517 74.3988 0.8324 73.9756 0.7186 76.1954 0.5778 95.9995 0.0137 96.1806 0.0137 96.2137
VFP 6050 0.8514 74.5476 0.8325 74.3897 0.7471 73.9474 0.2327 76.9317 0.0125 96.4989 0.0121 96.6196 0.012 96.6489
LFP 6106 0.8514 74.5476 0.8514 74.3897 0.8305 73.9474 0.711 76.9317 0.5713 96.4989 0.012 96.6196 0.012 96.6489
VFP 6016 0.8511 74.5452 0.8313 74.3806 0.7386 73.92 0.2044 77.7805 0.0108 96.9454 0.0108 97.0232 0.0105 97.0491
LFP 6072 0.8511 74.5452 0.8511 74.3806 0.8286 73.92 0.7038 77.7805 0.5652 96.9454 0.0104 97.0232 0.0104 97.0491
VFP 5982 0.8509 74.5429 0.8301 74.3716 0.7297 73.8938 0.1787 78.7341 0.0093 97.3312 0.0091 97.3912 0.009 97.4139
LFP 6038 0.8509 74.5429 0.8509 74.3716 0.8267 73.8938 0.6971 78.7341 0.5595 97.3312 0.009 97.3912 0.009 97.4139
24
25
18
19
20
21
22
23
12
13
14
15
16
17
6
7
8
9
10
11
Reboiler
1
2
3
4
5
Total Con. Part. Cond. 2 Top Tray Middle Tray Bottom Tray Part. Cond. 1
142
APPENDIX C: Miscellaneous Figures/Tables
Figure C1. Example Chromatograph Results
143
BIBLIOGRAPHY
144
BIBLIOGRAPHY
1) Cannon MR. Controlled Cycling Improves Various Processes. Ind Eng Chem.
1961;53:629-629.
2) Gaska RA, Cannon MR. Controlled Cycling Distillation in Sieve and Screen Plate
Towers. Ind Eng Chem. 1961;52:630-631.
3) McWhirter JR, Cannon MR. Controlled Cycling Distillation in a Packed-Plate Column.
Ind Eng Chem. 1961;52:632-634.
4) Schrodt VN. Controlled Cycle Operations. Ind. Engng Chem. Fundls. 1965;4:108-110.
5) Sommferfeld JT, Schrodt VN, Parisot PE, Chien HH. Studies of Controlled Cyclic
Distillation: I. Computer Simulations and the Analogy with Conventional Operation.
Separat. Sci. 1966;1:245-279.
6) Chien HH, Sommferfeld JT, Schrodt VN, Parisot PE. Studies of Controlled Cyclic
Distillation: II. Analytical Transient Solution and Asymptotic Plate Efficiencies. Separat.
Sci. 1966;1:281-317.
7) Schrodt VN, Sommerfeld JT, Martin OP, Parisot PE, Chien HH. Plant-scale study of
controlled cyclic distillation. Chem Eng Sci. 1966;22:759-767.
8) Horn FJM. Periodic Countercurrent Processes. I&EC Process Design and Develop.
1966;6:30-35.
9) Robinson RG, Engel AJ. An Analysis of Controlled Cycling Mass Transfer Operations.
Ind Eng Chem. 1967;59:22-29.
10) May RA, Horn FJM. Stage Efficiency of a Periodically Operated Distillation Column.
I&EC Process Design and Develop. 1968;7:61-64
11) May RA, Horn FJM. Effect of Mixing on Periodic Countercurrent Processes. I&EC
Process Design and Develop. 1968;7:349-354.
12) Furzer IA. Periodic Cycling of Plate Columns. Chem Eng Sci. 1973;28:296-299.
13) Gel’Perin NI, Polotskii LM, Potapov TG. Operation of a Bubble-Cap Fractionating
Column in a Cyclic Regime. Chem Petro Eng. 1976;11:707-709.
14) Rivas OR. An Analytical Solution of Cyclic Mass Transfer Operations. Ind Eng Chem,
Process Des Dev. 1977;16:400-402.
145
15) Duffy GJ, Furzer IA. Periodic Cycling of Plate Columns. Chem Eng Sci. 1978;33:897-
904
16) Furzer IA. The Discrete Residence Time Distribution of a Distillation Column Operated
with Microprocessor Controlled Periodic Cycling. Canadian Journal of Chem Eng.
1978;56:747-750.
17) Goss DW, Furzer IA. Mass Transfer in Periodically Cycled Plate Columns Containing
Multiple Sieve Plates. AIChe Journal. 1980;26:663-669.
18) Baron G, Wajc S, Lavie R. Stepwise Periodic Distillation – I: Total Reflux Operation.
Chem Eng Sci. 1980.;35:859-865
19) Baron G, Wajc S, Lavie R. Stepwise Periodic Distillation – II: Separation of a Binary
Mixture. Chem Eng Sci. 1981;36:1819-1827.
20) Matsubara M, Watanabe N, Kurimoto H. Binary Periodic Distillation Scheme with
Enhanced Energy Conservation. Chem Eng Sci. 1985;40:715-721.
21) Matsubara M, Watanabe N, Kurimoto H, Shimizu K. Binary Periodic Distillation Scheme
with Enhanced Energy Conservation - II. Chem Eng Sci. 1985;40:755-758.
22) Thompson MF, Furzer IA. Hydrodynamic Modelling for Liquid Holdups in Periodically
Cycled Plate Columns. AIChe Journal. 1984;30:496-500.
23) Szonyi L, Furzer IA. Periodic Cycling of Distillation Columns Using a New Tray Design.
AIChe Journal. 1985;31:1707-1713.
24) Bausa J, Tsatsaronis G. Reducing the Energy Demand of Continuous Distillation
Processes by Optimal Controlled Forced Periodic Operation. Comp Chem Eng.
2001;25:359-370.
25) Flodman HR, Timm DC. Batch Distillation Employing Cyclic Rectification and Stripping
Operations. ISA Trans. 2012;51:454-460.
26) Lita I, Bilde CS, Kiss AA. Modeling, Design and Control of Cyclic Distillation Systems.
Procedia Eng. 2012;42:1202-1213.
27) Maleta B, MaletaO. Mass Exchange Contact Device. US Patent 8,158,073, 2012.
28) Maleta BV, Shevchenko A, Bedryk O, Kiss AA. Pilot-Scale Studies of Process
Intensification by Cyclic Distillation.
29) Kiss AA. Distillation technology – still young and full of breakthrough opportunities. J
Chem Technol Biotechnol. 2014; 89:479-498.
146
30) Gerven TV, Stankiewicz A. Structure, Energy, Synergy, Time – The Fundamentals of
Process Intensification. Ind.Eng. Chem. Res. 2009;48:2465-2474.
31) White DC. Optimize Energy use in Distillation. Chem Eng. Prog. 2012;35-41.
32) Bildea CS, Patrut C, Jorgensen SB, Abildskov J, Kiss AA. Cyclic distillation technology
– a mini-review. J Chem Technol Biotechnol.2016;91:1215-1223.
33) Kiss AA. Cyclic Distillation. Advanced Distillation Technologies: Design, Control and
Applications. Chichester, West Sussex: Wiley, a John Wiley & Sons. 2013:311-351.
34) Waterbury M. 1,000 Craft Distilleries - and Counting! – Now operating in the USA.
Thewhiskeywash.com. 2016.
35) Distilled Spirits Council of the United States. Economic Contributions of Distilled Spirits
Industry. Discus.org/economics. 2016.
36) Boes RE, Berglund KA. Real Time Spirit Distillation Analysis using Gas
Chromatography with a Flow Vial Sampler.
37) Toftegard B, Jorgensen SB. An Integration Method for Dynamic Simulation of Cycled
Processes. Computers chem. Eng. 1989;13:927-930.
38) Sorensen E, Prenzler M. A Cyclic Operating Policy For Batch Distillation – Theory and
Practice. Computers chem. Eng. 1997;21:S1215-S1220.
39) Sorensen E. A Cyclic Operating Policy for Batch Distillation – Theory and Practice.
Computers chem. Eng. 1999.;23:533-542.
40) Distilled Spirits Council. “Distilled Spirits Council Reports Record Spirits Sales”
DistiledSpirits.org. February 1, 2018. https://www.distilledspirits.org/news/distilled-
spirits-council-reports-record-spirits-sales/
41) Hadrich B, Kechaou N. Identification of Best Model for Equilibrium Data of Ethanol-
Water Mixture. Journal of Chemistry and Chemical Engineering. 2010;4:46-48.
42) Seader J ,Henley E, RoperD. Separation Process Principals: Chemical and Biochemical
Operations.2011;3:481-483.
43) Claus M, Berglund K. Fruit Brandy Production By Batch Column Distillation With
Reflux. Journal of Food Processing Engineering.2004;28:53-67.
44) National Institute of Standards and Technology. NIST Chemistry WebBook. 2018.
https://webbook.nist.gov/chemistry/